Full-Output.txt

Drive already mounted at /content/gdrive; to attempt to forcibly remount, call drive.mount("/content/gdrive", force_remount=True).
Working in: /content/gdrive/MyDrive/Sapienza-Work-Place/SL_G21/Homework-1
/content/gdrive/MyDrive/Sapienza-Work-Place/SL_G21/Homework-1/linear-diffusion
Balanced test set contains {0: 50, 1: 50, 2: 50, 3: 50, 4: 50, 5: 50, 6: 50, 7: 50, 8: 50, 9: 50}
/usr/local/lib/python3.11/dist-packages/sklearn/linear_model/_sag.py:350: ConvergenceWarning: The max_iter was reached which means the coef_ did not converge
  warnings.warn(
LogisticRegression classifier trained.

=== Diffusion steps: T=400 ===

--- latent_size=8 | std=1.0 ---

== PCA [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.762
Confusion matrix (rows=true, cols=pred):
 [[31  0  1  2  0 14  0  0  2  0]
 [ 0 31  0  0  0  1  0  0 18  0]
 [ 0  0 37  3  2  2  1  0  4  1]
 [ 0  0  7 29  0  6  0  1  7  0]
 [ 0  0  0  0 44  1  0  0  2  3]
 [ 1  0  2  0  0 42  0  0  5  0]
 [ 0  1  0  0  0  5 41  0  3  0]
 [ 0  0  0  0  1  0  0 42  0  7]
 [ 0  0  0  2  0  6  0  0 41  1]
 [ 0  0  0  0  1  0  0  0  6 43]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.62      0.76        50
           1       0.97      0.62      0.76        50
           2       0.79      0.74      0.76        50
           3       0.81      0.58      0.67        50
           4       0.92      0.88      0.90        50
           5       0.55      0.84      0.66        50
           6       0.98      0.82      0.89        50
           7       0.98      0.84      0.90        50
           8       0.47      0.82      0.59        50
           9       0.78      0.86      0.82        50

    accuracy                           0.76       500
   macro avg       0.82      0.76      0.77       500
weighted avg       0.82      0.76      0.77       500

Generation time: 0.5937 seconds
Coverage=0.582
Avg set size=0.69
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.370    | 0.56     | 0.40    | 1.74     | 0.46    | 0.22    | 0.12    | 0      
1     | 50   | 0.362    | 0.56     | 0.36    | 1.46     | 0.45    | 0.24    | 0.12    | 1      
2     | 50   | 0.479    | 0.72     | 0.62    | 1.50     | 0.54    | 0.16    | 0.10    | 2      
3     | 50   | 0.376    | 0.54     | 0.36    | 1.86     | 0.47    | 0.22    | 0.12    | 3      
4     | 50   | 0.532    | 0.76     | 0.70    | 1.22     | 0.56    | 0.18    | 0.11    | 4      
5     | 50   | 0.525    | 0.68     | 0.60    | 1.22     | 0.55    | 0.19    | 0.09    | 5      
6     | 50   | 0.497    | 0.72     | 0.68    | 1.26     | 0.52    | 0.18    | 0.10    | 6      
7     | 50   | 0.546    | 0.84     | 0.76    | 1.20     | 0.58    | 0.22    | 0.09    | 7      
8     | 50   | 0.498    | 0.70     | 0.58    | 1.18     | 0.53    | 0.20    | 0.09    | 8      
9     | 50   | 0.547    | 0.78     | 0.76    | 1.28     | 0.57    | 0.18    | 0.10    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.898
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  3 40  0  0  0  0  7  0]
 [ 0  0  0  0 33  0  0  0  4 13]
 [ 0  0  0  0  0 44  0  0  6  0]
 [ 0  0  2  0  0  0 48  0  0  0]
 [ 0  0  0  0  0  0  0 35  5 10]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  1 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       0.91      1.00      0.95        50
           3       1.00      0.80      0.89        50
           4       1.00      0.66      0.80        50
           5       1.00      0.88      0.94        50
           6       1.00      0.96      0.98        50
           7       1.00      0.70      0.82        50
           8       0.68      1.00      0.81        50
           9       0.68      0.98      0.80        50

    accuracy                           0.90       500
   macro avg       0.93      0.90      0.90       500
weighted avg       0.93      0.90      0.90       500

Generation time: 1.2846 seconds
Coverage=0.538
Avg set size=0.54
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.539    | 0.94     | 0.94    | 1.00     | 0.54    | 0.20    | 0.11    | 0      
1     | 50   | 0.579    | 1.00     | 1.00    | 1.00     | 0.58    | 0.17    | 0.08    | 1      
2     | 50   | 0.491    | 0.82     | 0.82    | 1.00     | 0.49    | 0.20    | 0.09    | 2      
3     | 50   | 0.348    | 0.18     | 0.18    | 1.28     | 0.35    | 0.25    | 0.18    | 3      
4     | 50   | 0.302    | 0.02     | 0.02    | 1.38     | 0.31    | 0.26    | 0.18    | 4      
5     | 50   | 0.374    | 0.10     | 0.10    | 1.12     | 0.38    | 0.32    | 0.08    | 5      
6     | 50   | 0.381    | 0.32     | 0.32    | 1.04     | 0.38    | 0.18    | 0.14    | 6      
7     | 50   | 0.311    | 0.18     | 0.18    | 1.36     | 0.32    | 0.27    | 0.14    | 7      
8     | 50   | 0.602    | 1.00     | 1.00    | 1.00     | 0.60    | 0.12    | 0.08    | 8      
9     | 50   | 0.501    | 0.82     | 0.82    | 1.02     | 0.50    | 0.16    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.928
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  1 39  0  1  0  0  9  0]
 [ 0  0  0  0 42  0  0  0  5  3]
 [ 0  0  0  0  0 49  0  0  1  0]
 [ 0  0  3  0  0  0 47  0  0  0]
 [ 0  0  0  0  0  0  0 38  4  8]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  1 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       0.93      1.00      0.96        50
           3       1.00      0.78      0.88        50
           4       1.00      0.84      0.91        50
           5       0.98      0.98      0.98        50
           6       1.00      0.94      0.97        50
           7       1.00      0.76      0.86        50
           8       0.71      1.00      0.83        50
           9       0.82      0.98      0.89        50

    accuracy                           0.93       500
   macro avg       0.94      0.93      0.93       500
weighted avg       0.94      0.93      0.93       500

Generation time: 0.9362 seconds
Coverage=0.520
Avg set size=0.52
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.518    | 0.84     | 0.84    | 1.00     | 0.52    | 0.21    | 0.11    | 0      
1     | 50   | 0.739    | 1.00     | 1.00    | 1.00     | 0.74    | 0.11    | 0.06    | 1      
2     | 50   | 0.446    | 0.60     | 0.60    | 1.00     | 0.45    | 0.22    | 0.10    | 2      
3     | 50   | 0.336    | 0.08     | 0.08    | 1.36     | 0.35    | 0.25    | 0.19    | 3      
4     | 50   | 0.348    | 0.16     | 0.16    | 1.18     | 0.35    | 0.26    | 0.15    | 4      
5     | 50   | 0.387    | 0.18     | 0.18    | 1.02     | 0.39    | 0.31    | 0.08    | 5      
6     | 50   | 0.375    | 0.34     | 0.34    | 1.06     | 0.38    | 0.18    | 0.15    | 6      
7     | 50   | 0.316    | 0.18     | 0.18    | 1.46     | 0.33    | 0.27    | 0.13    | 7      
8     | 50   | 0.629    | 1.00     | 1.00    | 1.00     | 0.63    | 0.12    | 0.08    | 8      
9     | 50   | 0.528    | 0.82     | 0.82    | 1.02     | 0.53    | 0.15    | 0.11    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.726
Confusion matrix (rows=true, cols=pred):
 [[41  0  0  0  0  9  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 38  0  0  0  0  0 12  0]
 [ 0  0  0 18  0  2  0  0 30  0]
 [ 0  0  0  0 32  0  0  0 12  6]
 [ 0  0  0  0  0 37  0  0 13  0]
 [ 0  0  4  0  0  4 32  0 10  0]
 [ 0  0  0  0  0  0  0 15 15 20]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.82      0.90        50
           1       1.00      1.00      1.00        50
           2       0.90      0.76      0.83        50
           3       1.00      0.36      0.53        50
           4       1.00      0.64      0.78        50
           5       0.71      0.74      0.73        50
           6       1.00      0.64      0.78        50
           7       1.00      0.30      0.46        50
           8       0.35      1.00      0.52        50
           9       0.66      1.00      0.79        50

    accuracy                           0.73       500
   macro avg       0.86      0.73      0.73       500
weighted avg       0.86      0.73      0.73       500

Generation time: 0.9256 seconds
Coverage=0.298
Avg set size=0.30
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.329    | 0.12     | 0.12    | 1.26     | 0.34    | 0.24    | 0.17    | 0      
1     | 50   | 0.816    | 1.00     | 1.00    | 1.00     | 0.82    | 0.07    | 0.04    | 1      
2     | 50   | 0.297    | 0.00     | 0.00    | 1.24     | 0.30    | 0.27    | 0.15    | 2      
3     | 50   | 0.264    | 0.00     | 0.00    | 1.90     | 0.31    | 0.26    | 0.20    | 8      
4     | 50   | 0.293    | 0.08     | 0.08    | 1.52     | 0.31    | 0.24    | 0.17    | 4      
5     | 50   | 0.347    | 0.02     | 0.02    | 1.26     | 0.35    | 0.31    | 0.09    | 5      
6     | 50   | 0.269    | 0.06     | 0.06    | 1.70     | 0.28    | 0.20    | 0.18    | 6      
7     | 50   | 0.229    | 0.00     | 0.00    | 2.46     | 0.27    | 0.24    | 0.16    | 9      
8     | 50   | 0.587    | 0.98     | 0.98    | 1.00     | 0.59    | 0.13    | 0.08    | 8      
9     | 50   | 0.479    | 0.72     | 0.72    | 1.00     | 0.48    | 0.17    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.1028 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 7.98     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.990
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 50  0  0  0  0  0  0]
 [ 0  0  0  0 48  0  0  0  0  2]
 [ 0  0  0  0  0 49  0  0  1  0]
 [ 0  0  0  0  0  0 50  0  0  0]
 [ 0  0  0  0  0  0  0 48  0  2]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       1.00      1.00      1.00        50
           3       1.00      1.00      1.00        50
           4       1.00      0.96      0.98        50
           5       1.00      0.98      0.99        50
           6       1.00      1.00      1.00        50
           7       1.00      0.96      0.98        50
           8       0.98      1.00      0.99        50
           9       0.93      1.00      0.96        50

    accuracy                           0.99       500
   macro avg       0.99      0.99      0.99       500
weighted avg       0.99      0.99      0.99       500

Generation time: 1.0474 seconds
Coverage=0.950
Avg set size=0.96
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.838    | 0.92     | 0.92    | 1.00     | 0.84    | 0.09    | 0.03    | 0      
1     | 50   | 0.503    | 0.88     | 0.88    | 1.00     | 0.50    | 0.18    | 0.10    | 1      
2     | 50   | 0.936    | 1.00     | 1.00    | 1.00     | 0.94    | 0.03    | 0.02    | 2      
3     | 50   | 0.753    | 0.98     | 0.98    | 1.00     | 0.75    | 0.11    | 0.07    | 3      
4     | 50   | 0.547    | 0.86     | 0.82    | 1.04     | 0.55    | 0.24    | 0.08    | 4      
5     | 50   | 0.566    | 0.96     | 0.94    | 1.02     | 0.57    | 0.28    | 0.06    | 5      
6     | 50   | 0.875    | 1.00     | 1.00    | 1.00     | 0.87    | 0.05    | 0.02    | 6      
7     | 50   | 0.791    | 1.04     | 0.98    | 1.04     | 0.79    | 0.16    | 0.02    | 7      
8     | 50   | 0.734    | 0.98     | 0.98    | 1.00     | 0.73    | 0.10    | 0.06    | 8      
9     | 50   | 0.663    | 1.00     | 1.00    | 1.00     | 0.66    | 0.16    | 0.09    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=3.58354e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.6253 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=4.30212e-09): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.628
Confusion matrix (rows=true, cols=pred):
 [[43  0  0  0  0  5  0  0  2  0]
 [ 0 39  0  0  0  0  0  0 11  0]
 [ 0  0 38  3  0  3  3  0  3  0]
 [ 0  0  3 33  0  7  0  0  7  0]
 [ 0  0  0  0  0  0  0  0  1 49]
 [ 0  0  5  1  0 11  0  0 32  1]
 [ 0  0 10  0  0  4 25  0 11  0]
 [ 0  0  0  0  0  0  0 33  0 17]
 [ 0  0  1  4  0  0  0  0 45  0]
 [ 0  0  0  0  0  0  0  2  1 47]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.86      0.92        50
           1       1.00      0.78      0.88        50
           2       0.67      0.76      0.71        50
           3       0.80      0.66      0.73        50
           4       0.00      0.00      0.00        50
           5       0.37      0.22      0.27        50
           6       0.89      0.50      0.64        50
           7       0.94      0.66      0.78        50
           8       0.40      0.90      0.55        50
           9       0.41      0.94      0.57        50

    accuracy                           0.63       500
   macro avg       0.65      0.63      0.61       500
weighted avg       0.65      0.63      0.61       500

Generation time: 1.0996 seconds
Coverage=0.472
Avg set size=0.59
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.694    | 0.86     | 0.80    | 1.26     | 0.72    | 0.13    | 0.07    | 0      
1     | 50   | 0.373    | 0.26     | 0.26    | 1.22     | 0.38    | 0.27    | 0.10    | 1      
2     | 50   | 0.601    | 0.84     | 0.70    | 1.42     | 0.67    | 0.15    | 0.08    | 2      
3     | 50   | 0.501    | 0.64     | 0.56    | 1.98     | 0.59    | 0.18    | 0.10    | 3      
4     | 50   | 0.167    | 0.48     | 0.00    | 2.76     | 0.42    | 0.19    | 0.14    | 9      
5     | 50   | 0.251    | 0.16     | 0.12    | 2.30     | 0.37    | 0.25    | 0.12    | 8      
6     | 50   | 0.448    | 0.58     | 0.48    | 2.40     | 0.55    | 0.20    | 0.10    | 6      
7     | 50   | 0.536    | 0.94     | 0.64    | 1.44     | 0.63    | 0.24    | 0.06    | 7      
8     | 50   | 0.399    | 0.38     | 0.38    | 1.12     | 0.40    | 0.22    | 0.12    | 8      
9     | 50   | 0.562    | 0.80     | 0.78    | 1.08     | 0.56    | 0.21    | 0.10    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.590
Confusion matrix (rows=true, cols=pred):
 [[49  0  0  0  0  1  0  0  0  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 29  0  0  0  0 21  0]
 [ 0  0  0  0  0  0  0  0 45  5]
 [ 0  0  0  0  0  5  0  0 45  0]
 [ 0  0  1  0  0  0 44  0  5  0]
 [ 0  0  0  0  0  0  0 24  0 26]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  6 44]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.98      0.99        50
           1       0.00      0.00      0.00        50
           2       0.98      1.00      0.99        50
           3       1.00      0.58      0.73        50
           4       0.00      0.00      0.00        50
           5       0.83      0.10      0.18        50
           6       1.00      0.88      0.94        50
           7       1.00      0.48      0.65        50
           8       0.23      1.00      0.37        50
           9       0.59      0.88      0.70        50

    accuracy                           0.59       500
   macro avg       0.66      0.59      0.55       500
weighted avg       0.66      0.59      0.55       500

Generation time: 1.0001 seconds
Coverage=0.222
Avg set size=0.22
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.423    | 0.44     | 0.44    | 1.02     | 0.42    | 0.23    | 0.16    | 0      
1     | 50   | 0.131    | 0.00     | 0.00    | 2.42     | 0.33    | 0.13    | 0.12    | 8      
2     | 50   | 0.498    | 0.78     | 0.78    | 1.00     | 0.50    | 0.21    | 0.08    | 2      
3     | 50   | 0.294    | 0.00     | 0.00    | 1.42     | 0.31    | 0.27    | 0.17    | 3      
4     | 50   | 0.169    | 0.00     | 0.00    | 3.00     | 0.26    | 0.22    | 0.17    | 8      
5     | 50   | 0.301    | 0.00     | 0.00    | 1.90     | 0.36    | 0.30    | 0.08    | 8      
6     | 50   | 0.299    | 0.00     | 0.00    | 1.14     | 0.31    | 0.21    | 0.17    | 6      
7     | 50   | 0.287    | 0.00     | 0.00    | 1.56     | 0.31    | 0.27    | 0.16    | 9      
8     | 50   | 0.505    | 1.00     | 1.00    | 1.00     | 0.51    | 0.14    | 0.10    | 8      
9     | 50   | 0.332    | 0.00     | 0.00    | 1.12     | 0.33    | 0.24    | 0.12    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9190 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.069    | 0.00     | 0.00    | 5.60     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.038    | 0.00     | 0.00    | 9.00     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.162    | 0.00     | 0.00    | 2.10     | 0.34    | 0.16    | 0.15    | 8      
3     | 50   | 0.117    | 0.00     | 0.00    | 3.50     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.02     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.184    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.10    | 8      
6     | 50   | 0.072    | 0.00     | 0.00    | 5.64     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 6.00     | 0.32    | 0.15    | 0.14    | 8      
8     | 50   | 0.373    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.141    | 0.00     | 0.00    | 2.84     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=8 | std=1.5 ---

== PCA [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.642
Confusion matrix (rows=true, cols=pred):
 [[24  0  3  3  0 14  0  2  3  1]
 [ 0 27  1  2  0  1  1  0 18  0]
 [ 0  0 36  5  2  2  1  0  4  0]
 [ 0  0  9 26  1  7  0  2  5  0]
 [ 0  0  1  0 38  3  2  0  3  3]
 [ 1  0  4  2  1 33  0  0  7  2]
 [ 0  1  2  1  3  5 34  0  4  0]
 [ 0  0  1  0  4  0  0 40  0  5]
 [ 0  1  1  6  0 12  0  0 28  2]
 [ 0  0  1  0  3  3  0  6  2 35]]
Classification report:
               precision    recall  f1-score   support

           0       0.96      0.48      0.64        50
           1       0.93      0.54      0.68        50
           2       0.61      0.72      0.66        50
           3       0.58      0.52      0.55        50
           4       0.73      0.76      0.75        50
           5       0.41      0.66      0.51        50
           6       0.89      0.68      0.77        50
           7       0.80      0.80      0.80        50
           8       0.38      0.56      0.45        50
           9       0.73      0.70      0.71        50

    accuracy                           0.64       500
   macro avg       0.70      0.64      0.65       500
weighted avg       0.70      0.64      0.65       500

Generation time: 0.8036 seconds
Coverage=0.528
Avg set size=0.72
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.346    | 0.66     | 0.38    | 2.10     | 0.51    | 0.22    | 0.11    | 0      
1     | 50   | 0.344    | 0.66     | 0.38    | 1.92     | 0.49    | 0.22    | 0.12    | 1      
2     | 50   | 0.515    | 0.78     | 0.62    | 1.78     | 0.62    | 0.17    | 0.08    | 2      
3     | 50   | 0.370    | 0.68     | 0.38    | 2.14     | 0.55    | 0.20    | 0.11    | 3      
4     | 50   | 0.484    | 0.72     | 0.58    | 1.52     | 0.56    | 0.19    | 0.11    | 4      
5     | 50   | 0.512    | 0.70     | 0.56    | 1.60     | 0.59    | 0.18    | 0.09    | 5      
6     | 50   | 0.452    | 0.68     | 0.54    | 1.66     | 0.54    | 0.20    | 0.11    | 6      
7     | 50   | 0.550    | 0.78     | 0.68    | 1.32     | 0.59    | 0.20    | 0.09    | 7      
8     | 50   | 0.428    | 0.74     | 0.48    | 1.66     | 0.56    | 0.19    | 0.10    | 8      
9     | 50   | 0.504    | 0.82     | 0.68    | 1.64     | 0.57    | 0.20    | 0.09    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.832
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  4 35  0  2  0  0  9  0]
 [ 0  0  0  0 28  0  0  0  8 14]
 [ 0  0  0  0  0 34  0  0 16  0]
 [ 0  0  5  0  0  0 43  0  2  0]
 [ 0  0  0  0  0  1  0 30  8 11]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  2 48]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      0.98      0.99        50
           2       0.84      0.98      0.91        50
           3       1.00      0.70      0.82        50
           4       1.00      0.56      0.72        50
           5       0.92      0.68      0.78        50
           6       1.00      0.86      0.92        50
           7       1.00      0.60      0.75        50
           8       0.52      1.00      0.68        50
           9       0.66      0.96      0.78        50

    accuracy                           0.83       500
   macro avg       0.89      0.83      0.84       500
weighted avg       0.89      0.83      0.84       500

Generation time: 0.9231 seconds
Coverage=0.522
Avg set size=0.53
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.534    | 0.78     | 0.78    | 1.00     | 0.53    | 0.20    | 0.12    | 0      
1     | 50   | 0.563    | 0.90     | 0.90    | 1.02     | 0.56    | 0.18    | 0.09    | 1      
2     | 50   | 0.483    | 0.76     | 0.76    | 1.02     | 0.48    | 0.20    | 0.10    | 2      
3     | 50   | 0.356    | 0.32     | 0.32    | 1.46     | 0.38    | 0.25    | 0.17    | 3      
4     | 50   | 0.299    | 0.10     | 0.10    | 1.60     | 0.32    | 0.25    | 0.16    | 4      
5     | 50   | 0.366    | 0.16     | 0.14    | 1.32     | 0.38    | 0.31    | 0.09    | 5      
6     | 50   | 0.374    | 0.32     | 0.32    | 1.22     | 0.39    | 0.18    | 0.14    | 6      
7     | 50   | 0.310    | 0.22     | 0.18    | 1.72     | 0.33    | 0.27    | 0.12    | 7      
8     | 50   | 0.584    | 0.96     | 0.96    | 1.00     | 0.58    | 0.13    | 0.09    | 8      
9     | 50   | 0.499    | 0.76     | 0.76    | 1.08     | 0.50    | 0.17    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.868
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  3 37  0  0  0  0 10  0]
 [ 0  0  0  0 37  0  0  0  9  4]
 [ 0  0  0  0  0 40  0  0 10  0]
 [ 0  0  5  0  0  1 42  0  2  0]
 [ 0  0  1  0  0  0  0 31  5 13]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  1  0  0  0  0  0  2 47]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       0.83      1.00      0.91        50
           3       1.00      0.74      0.85        50
           4       1.00      0.74      0.85        50
           5       0.98      0.80      0.88        50
           6       1.00      0.84      0.91        50
           7       1.00      0.62      0.77        50
           8       0.57      1.00      0.72        50
           9       0.73      0.94      0.82        50

    accuracy                           0.87       500
   macro avg       0.91      0.87      0.87       500
weighted avg       0.91      0.87      0.87       500

Generation time: 0.9491 seconds
Coverage=0.550
Avg set size=0.55
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.515    | 0.82     | 0.82    | 1.00     | 0.52    | 0.21    | 0.12    | 0      
1     | 50   | 0.720    | 1.00     | 1.00    | 1.00     | 0.72    | 0.12    | 0.06    | 1      
2     | 50   | 0.444    | 0.54     | 0.54    | 1.00     | 0.44    | 0.22    | 0.11    | 2      
3     | 50   | 0.341    | 0.26     | 0.26    | 1.46     | 0.37    | 0.25    | 0.18    | 3      
4     | 50   | 0.347    | 0.28     | 0.26    | 1.36     | 0.37    | 0.26    | 0.14    | 4      
5     | 50   | 0.379    | 0.24     | 0.24    | 1.20     | 0.38    | 0.30    | 0.09    | 5      
6     | 50   | 0.364    | 0.36     | 0.36    | 1.30     | 0.38    | 0.18    | 0.14    | 6      
7     | 50   | 0.320    | 0.30     | 0.28    | 1.80     | 0.34    | 0.27    | 0.12    | 7      
8     | 50   | 0.613    | 0.96     | 0.96    | 1.00     | 0.61    | 0.12    | 0.09    | 8      
9     | 50   | 0.524    | 0.78     | 0.78    | 1.10     | 0.53    | 0.17    | 0.11    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.700
Confusion matrix (rows=true, cols=pred):
 [[37  0  0  0  0 10  0  0  3  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 33  0  0  1  0  0 16  0]
 [ 0  0  0 25  0  5  0  0 20  0]
 [ 0  0  0  0 28  0  0  0 15  7]
 [ 0  0  0  0  0 34  0  0 16  0]
 [ 0  0  5  0  0  6 27  0 12  0]
 [ 0  0  0  0  0  1  0 18 18 13]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  2 48]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.74      0.85        50
           1       1.00      1.00      1.00        50
           2       0.87      0.66      0.75        50
           3       1.00      0.50      0.67        50
           4       1.00      0.56      0.72        50
           5       0.60      0.68      0.64        50
           6       1.00      0.54      0.70        50
           7       1.00      0.36      0.53        50
           8       0.33      1.00      0.50        50
           9       0.71      0.96      0.81        50

    accuracy                           0.70       500
   macro avg       0.85      0.70      0.72       500
weighted avg       0.85      0.70      0.72       500

Generation time: 0.9338 seconds
Coverage=0.324
Avg set size=0.33
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.329    | 0.20     | 0.20    | 1.48     | 0.35    | 0.23    | 0.17    | 0      
1     | 50   | 0.793    | 0.98     | 0.98    | 1.00     | 0.79    | 0.08    | 0.05    | 1      
2     | 50   | 0.298    | 0.00     | 0.00    | 1.34     | 0.31    | 0.26    | 0.15    | 2      
3     | 50   | 0.274    | 0.08     | 0.08    | 1.88     | 0.32    | 0.25    | 0.19    | 3      
4     | 50   | 0.294    | 0.16     | 0.16    | 1.78     | 0.33    | 0.24    | 0.15    | 4      
5     | 50   | 0.345    | 0.10     | 0.10    | 1.32     | 0.36    | 0.30    | 0.09    | 5      
6     | 50   | 0.264    | 0.12     | 0.12    | 2.12     | 0.30    | 0.20    | 0.17    | 6      
7     | 50   | 0.239    | 0.12     | 0.10    | 2.64     | 0.29    | 0.24    | 0.14    | 7      
8     | 50   | 0.571    | 0.88     | 0.88    | 1.00     | 0.57    | 0.14    | 0.09    | 8      
9     | 50   | 0.471    | 0.62     | 0.62    | 1.06     | 0.47    | 0.17    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.8053 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.049    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 7.82     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.956
Confusion matrix (rows=true, cols=pred):
 [[46  0  0  0  0  2  0  0  1  1]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  1 49  0  0  0  0  0  0]
 [ 0  0  0  0 41  0  0  0  0  9]
 [ 0  0  0  1  0 47  0  0  2  0]
 [ 0  0  0  0  0  0 50  0  0  0]
 [ 0  0  0  0  0  0  0 46  0  4]
 [ 0  0  1  0  0  0  0  0 49  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.92      0.96        50
           1       1.00      1.00      1.00        50
           2       0.96      1.00      0.98        50
           3       0.98      0.98      0.98        50
           4       1.00      0.82      0.90        50
           5       0.96      0.94      0.95        50
           6       1.00      1.00      1.00        50
           7       1.00      0.92      0.96        50
           8       0.94      0.98      0.96        50
           9       0.78      1.00      0.88        50

    accuracy                           0.96       500
   macro avg       0.96      0.96      0.96       500
weighted avg       0.96      0.96      0.96       500

Generation time: 1.2622 seconds
Coverage=0.902
Avg set size=0.92
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.782    | 0.92     | 0.90    | 1.18     | 0.79    | 0.11    | 0.04    | 0      
1     | 50   | 0.493    | 0.76     | 0.76    | 1.00     | 0.49    | 0.18    | 0.11    | 1      
2     | 50   | 0.916    | 1.00     | 1.00    | 1.00     | 0.92    | 0.04    | 0.02    | 2      
3     | 50   | 0.733    | 0.96     | 0.96    | 1.02     | 0.74    | 0.12    | 0.07    | 3      
4     | 50   | 0.547    | 0.76     | 0.70    | 1.18     | 0.56    | 0.22    | 0.08    | 4      
5     | 50   | 0.560    | 0.82     | 0.80    | 1.06     | 0.56    | 0.26    | 0.07    | 5      
6     | 50   | 0.856    | 0.98     | 0.98    | 1.00     | 0.86    | 0.06    | 0.03    | 6      
7     | 50   | 0.781    | 1.06     | 0.96    | 1.08     | 0.79    | 0.16    | 0.02    | 7      
8     | 50   | 0.707    | 0.98     | 0.96    | 1.02     | 0.71    | 0.12    | 0.07    | 8      
9     | 50   | 0.655    | 1.00     | 1.00    | 1.00     | 0.65    | 0.17    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=3.58354e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0432 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=4.30212e-09): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.554
Confusion matrix (rows=true, cols=pred):
 [[35  0  2  0  0  7  0  0  6  0]
 [ 0 37  0  0  0  0  0  0 13  0]
 [ 1  0 34  3  0  3  6  0  3  0]
 [ 0  0  3 28  0  7  4  0  8  0]
 [ 0  0  0  0  0  0  0  0  2 48]
 [ 0  0  9  2  0 13  0  1 23  2]
 [ 0  0  9  1  0 10 23  0  7  0]
 [ 0  1  0  1  0  0  0 30  3 15]
 [ 0  0  3 10  0  1  0  0 36  0]
 [ 0  0  0  3  0  0  0  5  1 41]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.70      0.81        50
           1       0.97      0.74      0.84        50
           2       0.57      0.68      0.62        50
           3       0.58      0.56      0.57        50
           4       0.00      0.00      0.00        50
           5       0.32      0.26      0.29        50
           6       0.70      0.46      0.55        50
           7       0.83      0.60      0.70        50
           8       0.35      0.72      0.47        50
           9       0.39      0.82      0.53        50

    accuracy                           0.55       500
   macro avg       0.57      0.55      0.54       500
weighted avg       0.57      0.55      0.54       500

Generation time: 1.0202 seconds
Coverage=0.452
Avg set size=0.63
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.624    | 0.78     | 0.66    | 1.86     | 0.70    | 0.14    | 0.07    | 0      
1     | 50   | 0.376    | 0.34     | 0.34    | 1.26     | 0.40    | 0.25    | 0.11    | 1      
2     | 50   | 0.557    | 0.86     | 0.64    | 1.80     | 0.70    | 0.14    | 0.07    | 2      
3     | 50   | 0.500    | 0.76     | 0.54    | 2.76     | 0.63    | 0.17    | 0.08    | 3      
4     | 50   | 0.165    | 0.44     | 0.00    | 3.00     | 0.41    | 0.21    | 0.14    | 9      
5     | 50   | 0.257    | 0.32     | 0.18    | 2.70     | 0.40    | 0.24    | 0.12    | 8      
6     | 50   | 0.429    | 0.66     | 0.46    | 3.10     | 0.60    | 0.18    | 0.10    | 6      
7     | 50   | 0.517    | 0.96     | 0.60    | 1.76     | 0.65    | 0.21    | 0.07    | 7      
8     | 50   | 0.391    | 0.40     | 0.38    | 1.34     | 0.42    | 0.23    | 0.12    | 8      
9     | 50   | 0.558    | 0.78     | 0.72    | 1.26     | 0.58    | 0.20    | 0.09    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.566
Confusion matrix (rows=true, cols=pred):
 [[42  0  1  0  0  5  0  0  2  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 48  0  0  0  0  0  2  0]
 [ 0  0  0 29  0  0  0  0 21  0]
 [ 0  0  0  0  2  0  0  0 41  7]
 [ 0  0  0  0  0  7  0  0 43  0]
 [ 0  0  2  0  0  0 40  0  8  0]
 [ 0  0  0  0  0  0  0 24  1 25]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  9 41]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.84      0.91        50
           1       0.00      0.00      0.00        50
           2       0.94      0.96      0.95        50
           3       1.00      0.58      0.73        50
           4       1.00      0.04      0.08        50
           5       0.58      0.14      0.23        50
           6       1.00      0.80      0.89        50
           7       1.00      0.48      0.65        50
           8       0.22      1.00      0.36        50
           9       0.56      0.82      0.67        50

    accuracy                           0.57       500
   macro avg       0.73      0.57      0.55       500
weighted avg       0.73      0.57      0.55       500

Generation time: 0.9341 seconds
Coverage=0.230
Avg set size=0.24
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.436    | 0.50     | 0.50    | 1.22     | 0.44    | 0.22    | 0.15    | 0      
1     | 50   | 0.131    | 0.00     | 0.00    | 2.70     | 0.33    | 0.14    | 0.12    | 8      
2     | 50   | 0.496    | 0.74     | 0.74    | 1.04     | 0.50    | 0.21    | 0.09    | 2      
3     | 50   | 0.293    | 0.00     | 0.00    | 1.50     | 0.32    | 0.26    | 0.17    | 3      
4     | 50   | 0.169    | 0.00     | 0.00    | 2.90     | 0.26    | 0.21    | 0.17    | 8      
5     | 50   | 0.297    | 0.08     | 0.02    | 1.86     | 0.36    | 0.29    | 0.09    | 8      
6     | 50   | 0.302    | 0.04     | 0.04    | 1.26     | 0.32    | 0.21    | 0.16    | 6      
7     | 50   | 0.291    | 0.04     | 0.04    | 1.62     | 0.32    | 0.27    | 0.16    | 9      
8     | 50   | 0.498    | 0.92     | 0.92    | 1.00     | 0.50    | 0.15    | 0.10    | 8      
9     | 50   | 0.330    | 0.04     | 0.04    | 1.18     | 0.34    | 0.24    | 0.12    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9402 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.070    | 0.00     | 0.00    | 5.38     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.038    | 0.00     | 0.00    | 9.00     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.163    | 0.00     | 0.00    | 2.14     | 0.34    | 0.16    | 0.15    | 8      
3     | 50   | 0.117    | 0.00     | 0.00    | 3.50     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.04     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.184    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.10    | 8      
6     | 50   | 0.072    | 0.00     | 0.00    | 5.58     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 6.00     | 0.32    | 0.15    | 0.14    | 8      
8     | 50   | 0.374    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.141    | 0.00     | 0.00    | 2.78     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=8 | std=2.0 ---

== PCA [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.550
Confusion matrix (rows=true, cols=pred):
 [[19  1  3  3  1 15  0  2  5  1]
 [ 0 25  2  2  0  3  1  0 17  0]
 [ 0  0 34  5  3  3  1  1  3  0]
 [ 0  0 10 22  1  8  0  4  5  0]
 [ 0  0  3  0 28  7  2  2  5  3]
 [ 2  0  4  2  2 28  3  0  7  2]
 [ 0  0  3  3  4  7 29  0  4  0]
 [ 1  1  3  1  4  0  0 35  0  5]
 [ 1  1  3  6  0 12  0  0 25  2]
 [ 0  0  4  0  6  3  0  6  1 30]]
Classification report:
               precision    recall  f1-score   support

           0       0.83      0.38      0.52        50
           1       0.89      0.50      0.64        50
           2       0.49      0.68      0.57        50
           3       0.50      0.44      0.47        50
           4       0.57      0.56      0.57        50
           5       0.33      0.56      0.41        50
           6       0.81      0.58      0.67        50
           7       0.70      0.70      0.70        50
           8       0.35      0.50      0.41        50
           9       0.70      0.60      0.65        50

    accuracy                           0.55       500
   macro avg       0.62      0.55      0.56       500
weighted avg       0.62      0.55      0.56       500

Generation time: 0.6353 seconds
Coverage=0.464
Avg set size=0.76
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.319    | 0.76     | 0.32    | 2.44     | 0.56    | 0.20    | 0.10    | 0      
1     | 50   | 0.328    | 0.64     | 0.34    | 2.24     | 0.52    | 0.21    | 0.11    | 1      
2     | 50   | 0.535    | 0.86     | 0.62    | 1.90     | 0.68    | 0.16    | 0.07    | 2      
3     | 50   | 0.360    | 0.74     | 0.36    | 2.46     | 0.62    | 0.18    | 0.09    | 3      
4     | 50   | 0.429    | 0.74     | 0.44    | 1.94     | 0.56    | 0.20    | 0.10    | 4      
5     | 50   | 0.497    | 0.76     | 0.54    | 1.84     | 0.63    | 0.18    | 0.08    | 5      
6     | 50   | 0.415    | 0.78     | 0.50    | 1.94     | 0.56    | 0.22    | 0.10    | 6      
7     | 50   | 0.535    | 0.76     | 0.60    | 1.58     | 0.61    | 0.19    | 0.10    | 7      
8     | 50   | 0.376    | 0.78     | 0.40    | 2.08     | 0.60    | 0.19    | 0.09    | 8      
9     | 50   | 0.458    | 0.78     | 0.52    | 1.90     | 0.59    | 0.20    | 0.09    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.786
Confusion matrix (rows=true, cols=pred):
 [[47  0  0  0  0  1  0  0  2  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 47  0  0  0  0  0  3  0]
 [ 0  0  6 34  0  2  0  0  8  0]
 [ 1  0  0  0 24  0  0  0  9 16]
 [ 0  0  0  0  0 34  0  0 16  0]
 [ 0  0  9  0  0  0 37  0  4  0]
 [ 0  0  1  0  0  5  0 25  6 13]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  1  0  1  0  0  0  2 46]]
Classification report:
               precision    recall  f1-score   support

           0       0.98      0.94      0.96        50
           1       1.00      0.98      0.99        50
           2       0.73      0.94      0.82        50
           3       1.00      0.68      0.81        50
           4       0.96      0.48      0.64        50
           5       0.81      0.68      0.74        50
           6       1.00      0.74      0.85        50
           7       1.00      0.50      0.67        50
           8       0.50      1.00      0.66        50
           9       0.61      0.92      0.74        50

    accuracy                           0.79       500
   macro avg       0.86      0.79      0.79       500
weighted avg       0.86      0.79      0.79       500

Generation time: 0.9217 seconds
Coverage=0.484
Avg set size=0.50
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.527    | 0.70     | 0.70    | 1.08     | 0.53    | 0.20    | 0.12    | 0      
1     | 50   | 0.545    | 0.72     | 0.72    | 1.02     | 0.55    | 0.19    | 0.10    | 1      
2     | 50   | 0.473    | 0.62     | 0.62    | 1.06     | 0.48    | 0.20    | 0.10    | 2      
3     | 50   | 0.365    | 0.36     | 0.36    | 1.50     | 0.40    | 0.24    | 0.16    | 3      
4     | 50   | 0.294    | 0.16     | 0.14    | 1.84     | 0.33    | 0.24    | 0.15    | 4      
5     | 50   | 0.358    | 0.18     | 0.14    | 1.32     | 0.38    | 0.29    | 0.10    | 5      
6     | 50   | 0.366    | 0.34     | 0.32    | 1.46     | 0.39    | 0.18    | 0.14    | 6      
7     | 50   | 0.307    | 0.30     | 0.26    | 2.20     | 0.35    | 0.28    | 0.12    | 7      
8     | 50   | 0.563    | 0.84     | 0.84    | 1.00     | 0.56    | 0.14    | 0.10    | 8      
9     | 50   | 0.496    | 0.74     | 0.74    | 1.16     | 0.50    | 0.18    | 0.11    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.800
Confusion matrix (rows=true, cols=pred):
 [[46  0  0  0  0  2  0  0  2  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 48  0  0  0  0  0  2  0]
 [ 0  0  6 30  0  2  0  0 12  0]
 [ 0  0  0  0 33  2  0  0  8  7]
 [ 0  0  0  0  0 36  0  0 14  0]
 [ 0  0  6  0  0  3 36  0  5  0]
 [ 0  0  1  0  0  2  0 28  8 11]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  2  0  0  0  0  0  4 44]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.92      0.96        50
           1       1.00      0.98      0.99        50
           2       0.76      0.96      0.85        50
           3       1.00      0.60      0.75        50
           4       1.00      0.66      0.80        50
           5       0.77      0.72      0.74        50
           6       1.00      0.72      0.84        50
           7       1.00      0.56      0.72        50
           8       0.47      1.00      0.64        50
           9       0.71      0.88      0.79        50

    accuracy                           0.80       500
   macro avg       0.87      0.80      0.81       500
weighted avg       0.87      0.80      0.81       500

Generation time: 0.9414 seconds
Coverage=0.544
Avg set size=0.56
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.512    | 0.74     | 0.74    | 1.10     | 0.52    | 0.20    | 0.12    | 0      
1     | 50   | 0.697    | 0.96     | 0.96    | 1.02     | 0.70    | 0.13    | 0.07    | 1      
2     | 50   | 0.439    | 0.52     | 0.52    | 1.06     | 0.44    | 0.22    | 0.11    | 2      
3     | 50   | 0.347    | 0.30     | 0.28    | 1.70     | 0.39    | 0.24    | 0.16    | 3      
4     | 50   | 0.345    | 0.36     | 0.32    | 1.78     | 0.38    | 0.25    | 0.13    | 4      
5     | 50   | 0.371    | 0.28     | 0.26    | 1.28     | 0.38    | 0.29    | 0.10    | 5      
6     | 50   | 0.355    | 0.38     | 0.36    | 1.58     | 0.39    | 0.19    | 0.14    | 6      
7     | 50   | 0.323    | 0.38     | 0.34    | 2.10     | 0.36    | 0.27    | 0.12    | 7      
8     | 50   | 0.593    | 0.88     | 0.88    | 1.00     | 0.59    | 0.13    | 0.09    | 8      
9     | 50   | 0.517    | 0.78     | 0.78    | 1.30     | 0.53    | 0.17    | 0.10    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.666
Confusion matrix (rows=true, cols=pred):
 [[33  0  1  0  0 11  0  0  5  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 32  0  0  1  0  0 17  0]
 [ 0  0  2 27  0  5  0  0 16  0]
 [ 0  0  1  0 27  1  0  0 14  7]
 [ 0  0  0  0  0 31  0  0 19  0]
 [ 0  0  6  0  0  9 23  0 12  0]
 [ 0  0  1  0  0  1  0 19 18 11]
 [ 0  0  0  0  0  2  0  0 48  0]
 [ 0  0  0  0  0  0  0  0  6 44]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.66      0.80        50
           1       1.00      0.98      0.99        50
           2       0.74      0.64      0.69        50
           3       1.00      0.54      0.70        50
           4       1.00      0.54      0.70        50
           5       0.51      0.62      0.56        50
           6       1.00      0.46      0.63        50
           7       1.00      0.38      0.55        50
           8       0.31      0.96      0.47        50
           9       0.71      0.88      0.79        50

    accuracy                           0.67       500
   macro avg       0.83      0.67      0.69       500
weighted avg       0.83      0.67      0.69       500

Generation time: 1.5003 seconds
Coverage=0.344
Avg set size=0.35
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.329    | 0.26     | 0.26    | 1.68     | 0.36    | 0.23    | 0.16    | 0      
1     | 50   | 0.766    | 0.96     | 0.96    | 1.02     | 0.77    | 0.10    | 0.05    | 1      
2     | 50   | 0.298    | 0.00     | 0.00    | 1.40     | 0.32    | 0.26    | 0.14    | 2      
3     | 50   | 0.284    | 0.14     | 0.14    | 1.96     | 0.34    | 0.24    | 0.18    | 3      
4     | 50   | 0.296    | 0.26     | 0.24    | 2.04     | 0.35    | 0.24    | 0.14    | 4      
5     | 50   | 0.342    | 0.14     | 0.12    | 1.40     | 0.36    | 0.29    | 0.10    | 5      
6     | 50   | 0.262    | 0.20     | 0.20    | 2.42     | 0.32    | 0.20    | 0.16    | 6      
7     | 50   | 0.249    | 0.18     | 0.14    | 2.86     | 0.32    | 0.25    | 0.13    | 7      
8     | 50   | 0.552    | 0.78     | 0.78    | 1.04     | 0.55    | 0.15    | 0.10    | 8      
9     | 50   | 0.461    | 0.60     | 0.60    | 1.20     | 0.47    | 0.18    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.8164 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.089    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.043    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.049    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.042    | 0.00     | 0.00    | 7.64     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.916
Confusion matrix (rows=true, cols=pred):
 [[42  0  1  0  0  4  1  0  1  1]
 [ 0 47  0  0  0  1  0  0  2  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  1 48  0  0  0  0  1  0]
 [ 0  0  0  0 37  0  1  0  0 12]
 [ 0  1  0  2  0 44  0  0  3  0]
 [ 0  0  0  0  0  0 49  0  1  0]
 [ 0  0  0  0  0  0  0 44  0  6]
 [ 0  0  2  0  0  0  0  0 48  0]
 [ 0  0  0  0  0  0  0  1  0 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.84      0.91        50
           1       0.98      0.94      0.96        50
           2       0.93      1.00      0.96        50
           3       0.96      0.96      0.96        50
           4       1.00      0.74      0.85        50
           5       0.90      0.88      0.89        50
           6       0.96      0.98      0.97        50
           7       0.98      0.88      0.93        50
           8       0.86      0.96      0.91        50
           9       0.72      0.98      0.83        50

    accuracy                           0.92       500
   macro avg       0.93      0.92      0.92       500
weighted avg       0.93      0.92      0.92       500

Generation time: 0.8306 seconds
Coverage=0.836
Avg set size=0.88
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.730    | 0.84     | 0.80    | 1.54     | 0.76    | 0.12    | 0.05    | 0      
1     | 50   | 0.483    | 0.66     | 0.66    | 1.06     | 0.49    | 0.18    | 0.12    | 1      
2     | 50   | 0.886    | 0.98     | 0.98    | 1.00     | 0.89    | 0.06    | 0.02    | 2      
3     | 50   | 0.712    | 0.88     | 0.86    | 1.08     | 0.72    | 0.13    | 0.07    | 3      
4     | 50   | 0.536    | 0.78     | 0.68    | 1.30     | 0.57    | 0.21    | 0.08    | 4      
5     | 50   | 0.551    | 0.80     | 0.74    | 1.12     | 0.57    | 0.24    | 0.08    | 5      
6     | 50   | 0.835    | 0.94     | 0.94    | 1.04     | 0.84    | 0.08    | 0.04    | 6      
7     | 50   | 0.769    | 1.02     | 0.90    | 1.12     | 0.79    | 0.15    | 0.03    | 7      
8     | 50   | 0.676    | 0.96     | 0.92    | 1.08     | 0.69    | 0.13    | 0.07    | 8      
9     | 50   | 0.637    | 0.92     | 0.88    | 1.02     | 0.64    | 0.18    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=3.58354e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0419 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=4.30212e-09): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.522
Confusion matrix (rows=true, cols=pred):
 [[31  0  5  1  0  9  0  0  4  0]
 [ 0 35  0  0  0  0  0  0 15  0]
 [ 1  0 32  3  1  6  5  0  2  0]
 [ 0  1  3 27  0  6  7  0  5  1]
 [ 0  0  0  0  3  1  2  0  3 41]
 [ 0  0 12  3  0 15  0  3 16  1]
 [ 1  0  7  5  0  7 23  0  7  0]
 [ 0  1  0  5  0  0  0 26  5 13]
 [ 0  0  3 11  0  3  0  0 31  2]
 [ 0  0  1  5  0  0  0  4  2 38]]
Classification report:
               precision    recall  f1-score   support

           0       0.94      0.62      0.75        50
           1       0.95      0.70      0.80        50
           2       0.51      0.64      0.57        50
           3       0.45      0.54      0.49        50
           4       0.75      0.06      0.11        50
           5       0.32      0.30      0.31        50
           6       0.62      0.46      0.53        50
           7       0.79      0.52      0.63        50
           8       0.34      0.62      0.44        50
           9       0.40      0.76      0.52        50

    accuracy                           0.52       500
   macro avg       0.61      0.52      0.51       500
weighted avg       0.61      0.52      0.51       500

Generation time: 1.0334 seconds
Coverage=0.430
Avg set size=0.67
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.579    | 0.74     | 0.60    | 2.48     | 0.69    | 0.14    | 0.06    | 0      
1     | 50   | 0.379    | 0.38     | 0.38    | 1.32     | 0.41    | 0.23    | 0.11    | 1      
2     | 50   | 0.529    | 0.84     | 0.58    | 2.14     | 0.72    | 0.14    | 0.06    | 2      
3     | 50   | 0.494    | 0.80     | 0.52    | 3.04     | 0.67    | 0.16    | 0.07    | 3      
4     | 50   | 0.162    | 0.40     | 0.00    | 3.12     | 0.40    | 0.22    | 0.13    | 9      
5     | 50   | 0.266    | 0.46     | 0.24    | 2.86     | 0.45    | 0.22    | 0.12    | 8      
6     | 50   | 0.414    | 0.78     | 0.42    | 3.50     | 0.64    | 0.18    | 0.09    | 6      
7     | 50   | 0.491    | 0.88     | 0.50    | 2.08     | 0.67    | 0.20    | 0.06    | 7      
8     | 50   | 0.376    | 0.54     | 0.38    | 1.50     | 0.45    | 0.23    | 0.12    | 8      
9     | 50   | 0.543    | 0.84     | 0.68    | 1.50     | 0.61    | 0.18    | 0.09    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.554
Confusion matrix (rows=true, cols=pred):
 [[39  0  1  0  0  7  0  0  3  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 44  0  0  0  0  0  6  0]
 [ 0  0  1 29  0  0  0  0 20  0]
 [ 0  0  0  0  3  0  0  0 38  9]
 [ 0  0  0  0  0 10  0  0 40  0]
 [ 0  0  3  0  0  0 39  0  8  0]
 [ 0  0  0  1  0  0  0 24  5 20]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 11 39]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.78      0.88        50
           1       0.00      0.00      0.00        50
           2       0.90      0.88      0.89        50
           3       0.97      0.58      0.72        50
           4       1.00      0.06      0.11        50
           5       0.59      0.20      0.30        50
           6       1.00      0.78      0.88        50
           7       1.00      0.48      0.65        50
           8       0.22      1.00      0.36        50
           9       0.57      0.78      0.66        50

    accuracy                           0.55       500
   macro avg       0.72      0.55      0.54       500
weighted avg       0.72      0.55      0.54       500

Generation time: 0.9336 seconds
Coverage=0.248
Avg set size=0.26
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.445    | 0.54     | 0.52    | 1.38     | 0.47    | 0.21    | 0.14    | 0      
1     | 50   | 0.130    | 0.00     | 0.00    | 2.80     | 0.33    | 0.14    | 0.12    | 8      
2     | 50   | 0.491    | 0.68     | 0.68    | 1.12     | 0.50    | 0.20    | 0.10    | 2      
3     | 50   | 0.295    | 0.00     | 0.00    | 1.56     | 0.33    | 0.25    | 0.17    | 3      
4     | 50   | 0.169    | 0.00     | 0.00    | 2.86     | 0.27    | 0.21    | 0.16    | 8      
5     | 50   | 0.293    | 0.16     | 0.04    | 1.80     | 0.36    | 0.28    | 0.10    | 8      
6     | 50   | 0.304    | 0.20     | 0.18    | 1.30     | 0.33    | 0.21    | 0.15    | 6      
7     | 50   | 0.296    | 0.12     | 0.12    | 1.68     | 0.33    | 0.26    | 0.15    | 7      
8     | 50   | 0.489    | 0.86     | 0.86    | 1.00     | 0.49    | 0.16    | 0.10    | 8      
9     | 50   | 0.330    | 0.08     | 0.08    | 1.22     | 0.34    | 0.23    | 0.13    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9613 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.071    | 0.00     | 0.00    | 5.18     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.038    | 0.00     | 0.00    | 8.94     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.163    | 0.00     | 0.00    | 2.24     | 0.34    | 0.16    | 0.15    | 8      
3     | 50   | 0.117    | 0.00     | 0.00    | 3.48     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.06     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.184    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.10    | 8      
6     | 50   | 0.072    | 0.00     | 0.00    | 5.56     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.067    | 0.00     | 0.00    | 5.94     | 0.32    | 0.15    | 0.14    | 8      
8     | 50   | 0.374    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.141    | 0.00     | 0.00    | 2.78     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=10 | std=1.0 ---

== PCA [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.810
Confusion matrix (rows=true, cols=pred):
 [[43  0  1  0  0  5  1  0  0  0]
 [ 0 35  0  0  0  0  0  0 15  0]
 [ 0  0 31  1  0  5  0  2 10  1]
 [ 0  0  5 28  0 10  0  0  5  2]
 [ 0  0  1  0 46  1  0  0  0  2]
 [ 0  0  1  1  1 39  2  0  6  0]
 [ 0  0  1  0  0  0 48  0  1  0]
 [ 0  0  0  0  0  0  0 48  0  2]
 [ 0  0  1  2  1  1  0  0 43  2]
 [ 0  0  0  0  0  1  0  4  1 44]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.86      0.92        50
           1       1.00      0.70      0.82        50
           2       0.76      0.62      0.68        50
           3       0.88      0.56      0.68        50
           4       0.96      0.92      0.94        50
           5       0.63      0.78      0.70        50
           6       0.94      0.96      0.95        50
           7       0.89      0.96      0.92        50
           8       0.53      0.86      0.66        50
           9       0.83      0.88      0.85        50

    accuracy                           0.81       500
   macro avg       0.84      0.81      0.81       500
weighted avg       0.84      0.81      0.81       500

Generation time: 0.6351 seconds
Coverage=0.644
Avg set size=0.72
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.469    | 0.74     | 0.62    | 1.16     | 0.50    | 0.22    | 0.09    | 0      
1     | 50   | 0.446    | 0.60     | 0.56    | 1.32     | 0.48    | 0.24    | 0.11    | 1      
2     | 50   | 0.415    | 0.60     | 0.46    | 1.80     | 0.49    | 0.18    | 0.11    | 2      
3     | 50   | 0.380    | 0.50     | 0.36    | 1.90     | 0.48    | 0.20    | 0.12    | 3      
4     | 50   | 0.595    | 0.80     | 0.78    | 1.10     | 0.60    | 0.16    | 0.10    | 4      
5     | 50   | 0.527    | 0.74     | 0.68    | 1.28     | 0.55    | 0.21    | 0.08    | 5      
6     | 50   | 0.575    | 0.82     | 0.80    | 1.04     | 0.58    | 0.17    | 0.09    | 6      
7     | 50   | 0.624    | 0.96     | 0.88    | 1.04     | 0.63    | 0.19    | 0.07    | 7      
8     | 50   | 0.499    | 0.70     | 0.66    | 1.18     | 0.52    | 0.18    | 0.10    | 8      
9     | 50   | 0.488    | 0.72     | 0.64    | 1.18     | 0.51    | 0.20    | 0.10    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.872
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  1 35  0  4  0  0 10  0]
 [ 0  0  0  0 31  0  0  0  3 16]
 [ 0  0  0  0  0 38  0  0 12  0]
 [ 0  0  4  0  0  0 44  0  2  0]
 [ 0  0  0  0  0  0  0 38  5  7]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       0.91      1.00      0.95        50
           3       1.00      0.70      0.82        50
           4       1.00      0.62      0.77        50
           5       0.90      0.76      0.83        50
           6       1.00      0.88      0.94        50
           7       1.00      0.76      0.86        50
           8       0.61      1.00      0.76        50
           9       0.68      1.00      0.81        50

    accuracy                           0.87       500
   macro avg       0.91      0.87      0.87       500
weighted avg       0.91      0.87      0.87       500

Generation time: 1.1370 seconds
Coverage=0.524
Avg set size=0.52
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.543    | 0.84     | 0.84    | 1.00     | 0.54    | 0.20    | 0.11    | 0      
1     | 50   | 0.580    | 0.92     | 0.92    | 1.00     | 0.58    | 0.17    | 0.08    | 1      
2     | 50   | 0.492    | 0.88     | 0.88    | 1.00     | 0.49    | 0.20    | 0.09    | 2      
3     | 50   | 0.335    | 0.20     | 0.20    | 1.46     | 0.36    | 0.25    | 0.18    | 3      
4     | 50   | 0.311    | 0.06     | 0.06    | 1.42     | 0.33    | 0.26    | 0.17    | 4      
5     | 50   | 0.378    | 0.10     | 0.10    | 1.24     | 0.39    | 0.30    | 0.09    | 5      
6     | 50   | 0.379    | 0.42     | 0.42    | 1.16     | 0.39    | 0.18    | 0.14    | 6      
7     | 50   | 0.309    | 0.06     | 0.06    | 1.42     | 0.32    | 0.26    | 0.14    | 7      
8     | 50   | 0.591    | 0.98     | 0.98    | 1.00     | 0.59    | 0.13    | 0.09    | 8      
9     | 50   | 0.509    | 0.78     | 0.78    | 1.00     | 0.51    | 0.16    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.876
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 29  0  4  0  0 17  0]
 [ 0  0  0  0 38  0  0  0  5  7]
 [ 0  0  0  0  0 42  0  0  8  0]
 [ 0  0  3  0  0  1 44  0  2  0]
 [ 0  0  0  0  0  0  0 37  6  7]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  2 48]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       0.94      1.00      0.97        50
           3       1.00      0.58      0.73        50
           4       1.00      0.76      0.86        50
           5       0.89      0.84      0.87        50
           6       1.00      0.88      0.94        50
           7       1.00      0.74      0.85        50
           8       0.56      1.00      0.71        50
           9       0.77      0.96      0.86        50

    accuracy                           0.88       500
   macro avg       0.92      0.88      0.88       500
weighted avg       0.92      0.88      0.88       500

Generation time: 0.9840 seconds
Coverage=0.550
Avg set size=0.55
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.523    | 0.84     | 0.84    | 1.00     | 0.52    | 0.21    | 0.11    | 0      
1     | 50   | 0.741    | 1.00     | 1.00    | 1.00     | 0.74    | 0.10    | 0.05    | 1      
2     | 50   | 0.449    | 0.62     | 0.62    | 1.00     | 0.45    | 0.21    | 0.10    | 2      
3     | 50   | 0.320    | 0.16     | 0.16    | 1.66     | 0.35    | 0.24    | 0.18    | 3      
4     | 50   | 0.353    | 0.22     | 0.22    | 1.26     | 0.36    | 0.27    | 0.15    | 4      
5     | 50   | 0.389    | 0.28     | 0.28    | 1.16     | 0.39    | 0.30    | 0.09    | 5      
6     | 50   | 0.408    | 0.52     | 0.52    | 1.20     | 0.41    | 0.17    | 0.14    | 6      
7     | 50   | 0.300    | 0.08     | 0.08    | 1.46     | 0.31    | 0.25    | 0.14    | 7      
8     | 50   | 0.617    | 1.00     | 1.00    | 1.00     | 0.62    | 0.12    | 0.08    | 8      
9     | 50   | 0.529    | 0.78     | 0.78    | 1.04     | 0.53    | 0.16    | 0.12    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.734
Confusion matrix (rows=true, cols=pred):
 [[46  0  0  0  0  4  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 39  0  0  0  0  0 11  0]
 [ 0  0  0 17  0  2  0  0 31  0]
 [ 0  0  0  0 31  0  0  0 10  9]
 [ 0  0  0  0  0 35  0  0 15  0]
 [ 0  0  2  0  0  2 39  0  7  0]
 [ 0  0  0  0  0  2  0 15 19 14]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  5 45]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.92      0.96        50
           1       1.00      1.00      1.00        50
           2       0.95      0.78      0.86        50
           3       1.00      0.34      0.51        50
           4       1.00      0.62      0.77        50
           5       0.78      0.70      0.74        50
           6       1.00      0.78      0.88        50
           7       1.00      0.30      0.46        50
           8       0.34      1.00      0.51        50
           9       0.66      0.90      0.76        50

    accuracy                           0.73       500
   macro avg       0.87      0.73      0.74       500
weighted avg       0.87      0.73      0.74       500

Generation time: 0.9834 seconds
Coverage=0.326
Avg set size=0.33
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.376    | 0.30     | 0.30    | 1.08     | 0.38    | 0.23    | 0.16    | 0      
1     | 50   | 0.806    | 1.00     | 1.00    | 1.00     | 0.81    | 0.07    | 0.04    | 1      
2     | 50   | 0.308    | 0.00     | 0.00    | 1.24     | 0.32    | 0.26    | 0.14    | 2      
3     | 50   | 0.255    | 0.00     | 0.00    | 1.98     | 0.32    | 0.25    | 0.19    | 8      
4     | 50   | 0.315    | 0.16     | 0.16    | 1.52     | 0.33    | 0.25    | 0.16    | 4      
5     | 50   | 0.343    | 0.00     | 0.00    | 1.30     | 0.35    | 0.30    | 0.09    | 5      
6     | 50   | 0.300    | 0.10     | 0.10    | 1.48     | 0.31    | 0.19    | 0.17    | 6      
7     | 50   | 0.214    | 0.02     | 0.02    | 2.64     | 0.27    | 0.23    | 0.15    | 8      
8     | 50   | 0.587    | 1.00     | 1.00    | 1.00     | 0.59    | 0.13    | 0.09    | 8      
9     | 50   | 0.476    | 0.68     | 0.68    | 1.10     | 0.48    | 0.17    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.3508 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 7.92     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.992
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 50  0  0  0  0  0  0]
 [ 0  0  0  0 48  0  0  0  0  2]
 [ 0  0  0  0  0 48  0  0  2  0]
 [ 0  0  0  0  0  0 50  0  0  0]
 [ 0  0  0  0  0  0  0 50  0  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       1.00      1.00      1.00        50
           3       1.00      1.00      1.00        50
           4       1.00      0.96      0.98        50
           5       1.00      0.96      0.98        50
           6       1.00      1.00      1.00        50
           7       1.00      1.00      1.00        50
           8       0.96      1.00      0.98        50
           9       0.96      1.00      0.98        50

    accuracy                           0.99       500
   macro avg       0.99      0.99      0.99       500
weighted avg       0.99      0.99      0.99       500

Generation time: 1.3210 seconds
Coverage=0.974
Avg set size=0.98
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.833    | 0.96     | 0.96    | 1.00     | 0.83    | 0.08    | 0.04    | 0      
1     | 50   | 0.574    | 0.98     | 0.98    | 1.00     | 0.57    | 0.17    | 0.09    | 1      
2     | 50   | 0.894    | 0.98     | 0.98    | 1.00     | 0.89    | 0.05    | 0.03    | 2      
3     | 50   | 0.819    | 1.00     | 1.00    | 1.00     | 0.82    | 0.08    | 0.05    | 3      
4     | 50   | 0.585    | 0.92     | 0.92    | 1.04     | 0.59    | 0.23    | 0.07    | 4      
5     | 50   | 0.616    | 0.94     | 0.92    | 1.04     | 0.62    | 0.24    | 0.06    | 5      
6     | 50   | 0.853    | 0.98     | 0.98    | 1.00     | 0.85    | 0.06    | 0.03    | 6      
7     | 50   | 0.831    | 1.00     | 1.00    | 1.00     | 0.83    | 0.13    | 0.02    | 7      
8     | 50   | 0.719    | 1.00     | 1.00    | 1.00     | 0.72    | 0.12    | 0.06    | 8      
9     | 50   | 0.694    | 1.00     | 1.00    | 1.00     | 0.69    | 0.13    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=3.60332e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.4426 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=7.83564e-09): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.940
Confusion matrix (rows=true, cols=pred):
 [[38  0  0  0  0 10  1  0  1  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 49  0  0  0  0  1  0]
 [ 0  0  0  0 46  0  0  0  0  4]
 [ 0  0  0  1  1 41  0  0  7  0]
 [ 0  0  0  0  0  1 49  0  0  0]
 [ 0  0  0  0  0  0  0 47  0  3]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.76      0.86        50
           1       1.00      1.00      1.00        50
           2       1.00      1.00      1.00        50
           3       0.98      0.98      0.98        50
           4       0.98      0.92      0.95        50
           5       0.79      0.82      0.80        50
           6       0.98      0.98      0.98        50
           7       1.00      0.94      0.97        50
           8       0.85      1.00      0.92        50
           9       0.88      1.00      0.93        50

    accuracy                           0.94       500
   macro avg       0.95      0.94      0.94       500
weighted avg       0.95      0.94      0.94       500

Generation time: 1.0577 seconds
Coverage=0.918
Avg set size=0.95
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.688    | 0.82     | 0.70    | 1.48     | 0.75    | 0.13    | 0.06    | 0      
1     | 50   | 0.706    | 1.00     | 1.00    | 1.00     | 0.71    | 0.10    | 0.07    | 1      
2     | 50   | 0.876    | 0.98     | 0.98    | 1.00     | 0.88    | 0.06    | 0.02    | 2      
3     | 50   | 0.825    | 0.94     | 0.94    | 1.04     | 0.83    | 0.09    | 0.05    | 3      
4     | 50   | 0.597    | 0.92     | 0.86    | 1.08     | 0.60    | 0.24    | 0.07    | 4      
5     | 50   | 0.551    | 0.88     | 0.80    | 1.20     | 0.57    | 0.25    | 0.07    | 5      
6     | 50   | 0.902    | 0.98     | 0.98    | 1.02     | 0.90    | 0.04    | 0.02    | 6      
7     | 50   | 0.800    | 1.00     | 0.94    | 1.06     | 0.82    | 0.15    | 0.02    | 7      
8     | 50   | 0.694    | 0.98     | 0.98    | 1.00     | 0.69    | 0.15    | 0.06    | 8      
9     | 50   | 0.717    | 1.00     | 1.00    | 1.00     | 0.72    | 0.16    | 0.07    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.562
Confusion matrix (rows=true, cols=pred):
 [[39  0  0  0  0 10  0  0  1  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 27  0  0  0  0 23  0]
 [ 0  0  0  0  0  0  0  0 39 11]
 [ 0  0  0  0  0 11  0  0 39  0]
 [ 0  0  2  0  0  0 36  0 12  0]
 [ 0  0  0  0  0  0  0 20  0 30]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  2 48]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.78      0.88        50
           1       0.00      0.00      0.00        50
           2       0.96      1.00      0.98        50
           3       1.00      0.54      0.70        50
           4       0.00      0.00      0.00        50
           5       0.52      0.22      0.31        50
           6       1.00      0.72      0.84        50
           7       1.00      0.40      0.57        50
           8       0.23      1.00      0.38        50
           9       0.54      0.96      0.69        50

    accuracy                           0.56       500
   macro avg       0.63      0.56      0.53       500
weighted avg       0.63      0.56      0.53       500

Generation time: 0.9727 seconds
Coverage=0.220
Avg set size=0.22
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.402    | 0.44     | 0.44    | 1.32     | 0.42    | 0.23    | 0.15    | 0      
1     | 50   | 0.131    | 0.00     | 0.00    | 2.56     | 0.33    | 0.14    | 0.12    | 8      
2     | 50   | 0.466    | 0.72     | 0.72    | 1.00     | 0.47    | 0.22    | 0.09    | 2      
3     | 50   | 0.295    | 0.00     | 0.00    | 1.46     | 0.33    | 0.26    | 0.17    | 3      
4     | 50   | 0.166    | 0.00     | 0.00    | 2.98     | 0.26    | 0.22    | 0.17    | 8      
5     | 50   | 0.307    | 0.02     | 0.02    | 1.78     | 0.36    | 0.30    | 0.09    | 8      
6     | 50   | 0.284    | 0.00     | 0.00    | 1.32     | 0.30    | 0.20    | 0.17    | 6      
7     | 50   | 0.281    | 0.04     | 0.04    | 1.78     | 0.31    | 0.26    | 0.17    | 9      
8     | 50   | 0.499    | 0.94     | 0.94    | 1.00     | 0.50    | 0.15    | 0.10    | 8      
9     | 50   | 0.343    | 0.04     | 0.04    | 1.04     | 0.34    | 0.24    | 0.11    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9858 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.068    | 0.00     | 0.00    | 5.54     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.038    | 0.00     | 0.00    | 8.98     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.159    | 0.00     | 0.00    | 2.24     | 0.34    | 0.16    | 0.15    | 8      
3     | 50   | 0.116    | 0.00     | 0.00    | 3.58     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.00     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.185    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.10    | 8      
6     | 50   | 0.071    | 0.00     | 0.00    | 5.66     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.065    | 0.00     | 0.00    | 6.00     | 0.33    | 0.15    | 0.14    | 8      
8     | 50   | 0.373    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.142    | 0.00     | 0.00    | 2.72     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=10 | std=1.5 ---

== PCA [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.664
Confusion matrix (rows=true, cols=pred):
 [[33  0  4  1  0  9  1  0  2  0]
 [ 0 28  0  2  0  2  0  0 18  0]
 [ 0  1 28  1  1  8  1  3  7  0]
 [ 0  0  8 23  1 11  0  1  3  3]
 [ 0  0  1  0 39  6  0  0  1  3]
 [ 0  0  2  2  2 34  2  0  7  1]
 [ 1  0  2  0  2  4 38  0  3  0]
 [ 0  0  1  0  1  0  0 44  0  4]
 [ 0  1  1  3  4  7  1  0 30  3]
 [ 2  0  0  0  0  5  0  7  1 35]]
Classification report:
               precision    recall  f1-score   support

           0       0.92      0.66      0.77        50
           1       0.93      0.56      0.70        50
           2       0.60      0.56      0.58        50
           3       0.72      0.46      0.56        50
           4       0.78      0.78      0.78        50
           5       0.40      0.68      0.50        50
           6       0.88      0.76      0.82        50
           7       0.80      0.88      0.84        50
           8       0.42      0.60      0.49        50
           9       0.71      0.70      0.71        50

    accuracy                           0.66       500
   macro avg       0.72      0.66      0.67       500
weighted avg       0.72      0.66      0.67       500

Generation time: 0.6370 seconds
Coverage=0.558
Avg set size=0.71
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.412    | 0.64     | 0.46    | 1.42     | 0.50    | 0.23    | 0.10    | 0      
1     | 50   | 0.432    | 0.66     | 0.50    | 1.60     | 0.50    | 0.22    | 0.10    | 1      
2     | 50   | 0.427    | 0.72     | 0.48    | 2.16     | 0.57    | 0.18    | 0.10    | 2      
3     | 50   | 0.389    | 0.62     | 0.38    | 2.26     | 0.55    | 0.19    | 0.11    | 3      
4     | 50   | 0.552    | 0.80     | 0.70    | 1.40     | 0.60    | 0.16    | 0.10    | 4      
5     | 50   | 0.503    | 0.78     | 0.60    | 1.58     | 0.59    | 0.21    | 0.09    | 5      
6     | 50   | 0.534    | 0.80     | 0.70    | 1.32     | 0.57    | 0.20    | 0.10    | 6      
7     | 50   | 0.619    | 0.90     | 0.80    | 1.12     | 0.64    | 0.18    | 0.07    | 7      
8     | 50   | 0.421    | 0.62     | 0.48    | 1.68     | 0.51    | 0.21    | 0.11    | 8      
9     | 50   | 0.423    | 0.60     | 0.48    | 1.48     | 0.50    | 0.22    | 0.11    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.808
Confusion matrix (rows=true, cols=pred):
 [[48  0  0  0  0  1  0  0  1  0]
 [ 0 48  0  0  0  0  0  0  2  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  4 31  0  5  0  0 10  0]
 [ 0  0  0  0 26  0  0  0  6 18]
 [ 0  0  0  0  0 36  0  0 14  0]
 [ 0  0  9  0  0  2 37  0  2  0]
 [ 0  0  2  0  0  1  0 32  5 10]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  4 46]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.96      0.98        50
           1       1.00      0.96      0.98        50
           2       0.77      1.00      0.87        50
           3       1.00      0.62      0.77        50
           4       1.00      0.52      0.68        50
           5       0.80      0.72      0.76        50
           6       1.00      0.74      0.85        50
           7       1.00      0.64      0.78        50
           8       0.53      1.00      0.69        50
           9       0.62      0.92      0.74        50

    accuracy                           0.81       500
   macro avg       0.87      0.81      0.81       500
weighted avg       0.87      0.81      0.81       500

Generation time: 0.9873 seconds
Coverage=0.522
Avg set size=0.53
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.536    | 0.76     | 0.76    | 1.06     | 0.54    | 0.20    | 0.11    | 0      
1     | 50   | 0.567    | 0.82     | 0.82    | 1.04     | 0.57    | 0.17    | 0.09    | 1      
2     | 50   | 0.486    | 0.78     | 0.78    | 1.00     | 0.49    | 0.20    | 0.09    | 2      
3     | 50   | 0.337    | 0.34     | 0.34    | 1.80     | 0.38    | 0.24    | 0.17    | 3      
4     | 50   | 0.309    | 0.16     | 0.12    | 1.58     | 0.35    | 0.25    | 0.16    | 4      
5     | 50   | 0.368    | 0.20     | 0.18    | 1.28     | 0.39    | 0.29    | 0.10    | 5      
6     | 50   | 0.372    | 0.46     | 0.44    | 1.44     | 0.40    | 0.18    | 0.14    | 6      
7     | 50   | 0.303    | 0.16     | 0.14    | 1.70     | 0.33    | 0.26    | 0.13    | 7      
8     | 50   | 0.560    | 0.90     | 0.90    | 1.00     | 0.56    | 0.15    | 0.09    | 8      
9     | 50   | 0.508    | 0.74     | 0.74    | 1.08     | 0.51    | 0.16    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.838
Confusion matrix (rows=true, cols=pred):
 [[49  0  0  0  0  1  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  3 27  0  5  0  0 15  0]
 [ 0  0  0  0 34  0  0  0  6 10]
 [ 0  0  0  0  0 39  0  0 11  0]
 [ 0  0  6  0  0  1 41  0  2  0]
 [ 0  0  1  0  0  1  0 34  8  6]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  1  0  0  1  0  0  2 46]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.98      0.99        50
           1       1.00      1.00      1.00        50
           2       0.82      0.98      0.89        50
           3       1.00      0.54      0.70        50
           4       1.00      0.68      0.81        50
           5       0.81      0.78      0.80        50
           6       1.00      0.82      0.90        50
           7       1.00      0.68      0.81        50
           8       0.53      1.00      0.69        50
           9       0.74      0.92      0.82        50

    accuracy                           0.84       500
   macro avg       0.89      0.84      0.84       500
weighted avg       0.89      0.84      0.84       500

Generation time: 1.5263 seconds
Coverage=0.540
Avg set size=0.55
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.517    | 0.78     | 0.78    | 1.02     | 0.52    | 0.20    | 0.11    | 0      
1     | 50   | 0.720    | 0.96     | 0.96    | 1.00     | 0.72    | 0.11    | 0.06    | 1      
2     | 50   | 0.447    | 0.60     | 0.60    | 1.02     | 0.45    | 0.21    | 0.10    | 2      
3     | 50   | 0.319    | 0.30     | 0.26    | 2.00     | 0.37    | 0.23    | 0.17    | 3      
4     | 50   | 0.352    | 0.26     | 0.24    | 1.50     | 0.38    | 0.26    | 0.14    | 4      
5     | 50   | 0.379    | 0.30     | 0.30    | 1.22     | 0.39    | 0.28    | 0.10    | 5      
6     | 50   | 0.410    | 0.52     | 0.52    | 1.40     | 0.43    | 0.18    | 0.13    | 6      
7     | 50   | 0.299    | 0.14     | 0.14    | 1.82     | 0.33    | 0.25    | 0.13    | 7      
8     | 50   | 0.587    | 0.90     | 0.90    | 1.00     | 0.59    | 0.14    | 0.09    | 8      
9     | 50   | 0.522    | 0.70     | 0.70    | 1.18     | 0.53    | 0.16    | 0.11    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.700
Confusion matrix (rows=true, cols=pred):
 [[44  0  0  0  0  4  0  0  2  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 35  0  0  3  0  0 12  0]
 [ 0  0  0 20  0  5  0  0 25  0]
 [ 0  0  0  0 28  0  0  0 12 10]
 [ 0  0  0  0  0 32  0  0 18  0]
 [ 0  0  4  0  0  3 36  0  7  0]
 [ 0  0  0  0  0  4  0 16 17 13]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 10 40]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.88      0.94        50
           1       1.00      0.98      0.99        50
           2       0.90      0.70      0.79        50
           3       1.00      0.40      0.57        50
           4       1.00      0.56      0.72        50
           5       0.63      0.64      0.63        50
           6       1.00      0.72      0.84        50
           7       1.00      0.32      0.48        50
           8       0.32      1.00      0.49        50
           9       0.63      0.80      0.71        50

    accuracy                           0.70       500
   macro avg       0.85      0.70      0.72       500
weighted avg       0.85      0.70      0.72       500

Generation time: 0.9780 seconds
Coverage=0.362
Avg set size=0.36
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.383    | 0.34     | 0.34    | 1.22     | 0.39    | 0.23    | 0.15    | 0      
1     | 50   | 0.774    | 0.94     | 0.94    | 1.02     | 0.77    | 0.08    | 0.05    | 1      
2     | 50   | 0.312    | 0.06     | 0.06    | 1.36     | 0.33    | 0.25    | 0.14    | 2      
3     | 50   | 0.260    | 0.10     | 0.10    | 2.08     | 0.34    | 0.24    | 0.18    | 8      
4     | 50   | 0.321    | 0.26     | 0.26    | 1.84     | 0.36    | 0.25    | 0.14    | 4      
5     | 50   | 0.336    | 0.06     | 0.06    | 1.36     | 0.35    | 0.29    | 0.10    | 5      
6     | 50   | 0.313    | 0.24     | 0.24    | 1.70     | 0.34    | 0.19    | 0.16    | 6      
7     | 50   | 0.219    | 0.06     | 0.06    | 2.92     | 0.29    | 0.23    | 0.14    | 8      
8     | 50   | 0.566    | 0.92     | 0.92    | 1.00     | 0.57    | 0.14    | 0.10    | 8      
9     | 50   | 0.471    | 0.64     | 0.64    | 1.42     | 0.49    | 0.16    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.8603 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.049    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.042    | 0.00     | 0.00    | 7.72     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.966
Confusion matrix (rows=true, cols=pred):
 [[46  0  1  0  0  0  0  0  3  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  0 50  0  0  0  0  0  0]
 [ 0  0  0  0 46  0  0  0  0  4]
 [ 0  0  0  1  1 44  0  0  4  0]
 [ 0  0  0  0  0  0 49  0  1  0]
 [ 0  0  0  0  0  0  0 50  0  0]
 [ 0  0  0  1  0  0  0  0 49  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.92      0.96        50
           1       1.00      1.00      1.00        50
           2       0.98      0.98      0.98        50
           3       0.96      1.00      0.98        50
           4       0.98      0.92      0.95        50
           5       1.00      0.88      0.94        50
           6       1.00      0.98      0.99        50
           7       1.00      1.00      1.00        50
           8       0.84      0.98      0.91        50
           9       0.93      1.00      0.96        50

    accuracy                           0.97       500
   macro avg       0.97      0.97      0.97       500
weighted avg       0.97      0.97      0.97       500

Generation time: 1.0135 seconds
Coverage=0.912
Avg set size=0.93
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.753    | 0.86     | 0.86    | 1.16     | 0.77    | 0.11    | 0.05    | 0      
1     | 50   | 0.563    | 0.84     | 0.84    | 1.00     | 0.56    | 0.17    | 0.09    | 1      
2     | 50   | 0.855    | 0.98     | 0.98    | 1.04     | 0.86    | 0.07    | 0.03    | 2      
3     | 50   | 0.815    | 0.96     | 0.96    | 1.00     | 0.81    | 0.09    | 0.04    | 3      
4     | 50   | 0.568    | 0.86     | 0.80    | 1.08     | 0.58    | 0.21    | 0.08    | 4      
5     | 50   | 0.597    | 0.98     | 0.88    | 1.14     | 0.62    | 0.21    | 0.08    | 5      
6     | 50   | 0.815    | 0.96     | 0.96    | 1.06     | 0.82    | 0.08    | 0.04    | 6      
7     | 50   | 0.826    | 1.00     | 1.00    | 1.00     | 0.83    | 0.12    | 0.02    | 7      
8     | 50   | 0.662    | 0.92     | 0.90    | 1.02     | 0.67    | 0.15    | 0.07    | 8      
9     | 50   | 0.686    | 0.94     | 0.94    | 1.00     | 0.69    | 0.14    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=3.60332e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0937 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=7.83564e-09): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.866
Confusion matrix (rows=true, cols=pred):
 [[33  0  1  1  1  9  1  0  3  1]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 45  0  0  1  2  0  2  0]
 [ 0  0  2 45  0  0  0  1  2  0]
 [ 0  0  0  0 39  0  0  0  2  9]
 [ 0  0  0  3  1 39  1  0  4  2]
 [ 0  1  0  0  0  2 47  0  0  0]
 [ 0  0  0  0  0  0  0 44  1  5]
 [ 0  0  1  2  0  2  0  0 45  0]
 [ 0  0  0  0  2  0  0  2  0 46]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.66      0.80        50
           1       0.98      1.00      0.99        50
           2       0.92      0.90      0.91        50
           3       0.88      0.90      0.89        50
           4       0.91      0.78      0.84        50
           5       0.74      0.78      0.76        50
           6       0.92      0.94      0.93        50
           7       0.94      0.88      0.91        50
           8       0.76      0.90      0.83        50
           9       0.73      0.92      0.81        50

    accuracy                           0.87       500
   macro avg       0.88      0.87      0.87       500
weighted avg       0.88      0.87      0.87       500

Generation time: 1.1398 seconds
Coverage=0.840
Avg set size=0.94
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.600    | 0.88     | 0.64    | 2.24     | 0.74    | 0.14    | 0.05    | 0      
1     | 50   | 0.689    | 0.98     | 0.98    | 1.00     | 0.69    | 0.11    | 0.08    | 1      
2     | 50   | 0.804    | 0.94     | 0.88    | 1.24     | 0.82    | 0.09    | 0.03    | 2      
3     | 50   | 0.790    | 0.94     | 0.90    | 1.18     | 0.81    | 0.10    | 0.05    | 3      
4     | 50   | 0.549    | 0.88     | 0.70    | 1.28     | 0.58    | 0.23    | 0.08    | 4      
5     | 50   | 0.535    | 0.86     | 0.76    | 1.48     | 0.59    | 0.22    | 0.08    | 5      
6     | 50   | 0.840    | 0.96     | 0.92    | 1.16     | 0.86    | 0.06    | 0.03    | 6      
7     | 50   | 0.766    | 1.04     | 0.88    | 1.16     | 0.81    | 0.14    | 0.02    | 7      
8     | 50   | 0.628    | 0.94     | 0.84    | 1.12     | 0.64    | 0.19    | 0.08    | 8      
9     | 50   | 0.693    | 0.96     | 0.90    | 1.08     | 0.70    | 0.18    | 0.06    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.544
Confusion matrix (rows=true, cols=pred):
 [[38  0  0  0  0  8  0  0  4  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 46  0  0  0  0  0  4  0]
 [ 0  0  1 27  0  1  0  0 21  0]
 [ 0  0  0  0  1  0  0  0 35 14]
 [ 0  0  0  0  0 14  0  0 36  0]
 [ 0  0  5  0  0  0 31  0 14  0]
 [ 0  0  0  0  0  0  0 20  6 24]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  5 45]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.76      0.86        50
           1       0.00      0.00      0.00        50
           2       0.88      0.92      0.90        50
           3       1.00      0.54      0.70        50
           4       1.00      0.02      0.04        50
           5       0.61      0.28      0.38        50
           6       1.00      0.62      0.77        50
           7       1.00      0.40      0.57        50
           8       0.22      1.00      0.36        50
           9       0.54      0.90      0.68        50

    accuracy                           0.54       500
   macro avg       0.73      0.54      0.53       500
weighted avg       0.73      0.54      0.53       500

Generation time: 1.0948 seconds
Coverage=0.226
Avg set size=0.23
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.406    | 0.48     | 0.46    | 1.50     | 0.44    | 0.22    | 0.14    | 0      
1     | 50   | 0.131    | 0.00     | 0.00    | 2.92     | 0.33    | 0.14    | 0.12    | 8      
2     | 50   | 0.450    | 0.52     | 0.52    | 1.08     | 0.45    | 0.22    | 0.10    | 2      
3     | 50   | 0.298    | 0.08     | 0.06    | 1.52     | 0.34    | 0.24    | 0.17    | 3      
4     | 50   | 0.164    | 0.00     | 0.00    | 2.96     | 0.27    | 0.21    | 0.16    | 8      
5     | 50   | 0.303    | 0.04     | 0.02    | 1.72     | 0.36    | 0.28    | 0.10    | 8      
6     | 50   | 0.279    | 0.10     | 0.08    | 1.54     | 0.32    | 0.20    | 0.16    | 6      
7     | 50   | 0.282    | 0.08     | 0.08    | 1.82     | 0.32    | 0.26    | 0.16    | 9      
8     | 50   | 0.485    | 0.94     | 0.94    | 1.00     | 0.49    | 0.16    | 0.10    | 8      
9     | 50   | 0.344    | 0.10     | 0.10    | 1.10     | 0.35    | 0.23    | 0.12    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0898 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.068    | 0.00     | 0.00    | 5.52     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.039    | 0.00     | 0.00    | 8.80     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.158    | 0.00     | 0.00    | 2.34     | 0.34    | 0.16    | 0.15    | 8      
3     | 50   | 0.117    | 0.00     | 0.00    | 3.62     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.04     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.185    | 0.00     | 0.00    | 2.00     | 0.35    | 0.19    | 0.10    | 8      
6     | 50   | 0.071    | 0.00     | 0.00    | 5.62     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 5.94     | 0.33    | 0.15    | 0.14    | 8      
8     | 50   | 0.372    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.142    | 0.00     | 0.00    | 2.68     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=10 | std=2.0 ---

== PCA [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.542
Confusion matrix (rows=true, cols=pred):
 [[22  0  6  3  0 11  2  3  2  1]
 [ 0 26  2  2  0  4  0  0 16  0]
 [ 1  1 25  2  2  8  1  3  7  0]
 [ 0  0  9 22  1 11  0  1  3  3]
 [ 0  3  1  0 32  8  0  1  2  3]
 [ 0  0  4  2  3 30  3  0  7  1]
 [ 1  1  5  0  3  7 30  0  3  0]
 [ 0  1  3  0  2  0  0 39  0  5]
 [ 2  1  2  3  4 10  1  0 23  4]
 [ 2  0  0  2  1  7  0 12  4 22]]
Classification report:
               precision    recall  f1-score   support

           0       0.79      0.44      0.56        50
           1       0.79      0.52      0.63        50
           2       0.44      0.50      0.47        50
           3       0.61      0.44      0.51        50
           4       0.67      0.64      0.65        50
           5       0.31      0.60      0.41        50
           6       0.81      0.60      0.69        50
           7       0.66      0.78      0.72        50
           8       0.34      0.46      0.39        50
           9       0.56      0.44      0.49        50

    accuracy                           0.54       500
   macro avg       0.60      0.54      0.55       500
weighted avg       0.60      0.54      0.55       500

Generation time: 0.8660 seconds
Coverage=0.484
Avg set size=0.77
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.354    | 0.74     | 0.36    | 1.82     | 0.53    | 0.23    | 0.11    | 0      
1     | 50   | 0.411    | 0.70     | 0.48    | 1.92     | 0.52    | 0.22    | 0.10    | 1      
2     | 50   | 0.429    | 0.82     | 0.48    | 2.44     | 0.63    | 0.17    | 0.09    | 2      
3     | 50   | 0.393    | 0.72     | 0.38    | 2.54     | 0.60    | 0.18    | 0.09    | 3      
4     | 50   | 0.503    | 0.78     | 0.56    | 1.80     | 0.61    | 0.17    | 0.09    | 4      
5     | 50   | 0.486    | 0.88     | 0.60    | 1.88     | 0.61    | 0.21    | 0.08    | 5      
6     | 50   | 0.490    | 0.80     | 0.56    | 1.66     | 0.60    | 0.20    | 0.09    | 6      
7     | 50   | 0.595    | 0.94     | 0.74    | 1.24     | 0.64    | 0.19    | 0.08    | 7      
8     | 50   | 0.361    | 0.68     | 0.36    | 2.20     | 0.54    | 0.22    | 0.10    | 8      
9     | 50   | 0.360    | 0.66     | 0.32    | 1.96     | 0.52    | 0.22    | 0.11    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.752
Confusion matrix (rows=true, cols=pred):
 [[44  0  0  0  0  3  0  0  3  0]
 [ 0 44  0  0  0  0  0  0  6  0]
 [ 1  0 47  0  0  0  0  0  2  0]
 [ 1  0  7 29  0  6  0  0  7  0]
 [ 0  0  0  0 24  0  0  0  7 19]
 [ 1  0  0  0  0 32  0  0 17  0]
 [ 0  0 11  0  0  2 34  0  3  0]
 [ 1  0  1  0  0  3  0 28  7 10]
 [ 0  0  1  0  0  0  0  0 49  0]
 [ 0  0  0  0  0  1  0  0  4 45]]
Classification report:
               precision    recall  f1-score   support

           0       0.92      0.88      0.90        50
           1       1.00      0.88      0.94        50
           2       0.70      0.94      0.80        50
           3       1.00      0.58      0.73        50
           4       1.00      0.48      0.65        50
           5       0.68      0.64      0.66        50
           6       1.00      0.68      0.81        50
           7       1.00      0.56      0.72        50
           8       0.47      0.98      0.63        50
           9       0.61      0.90      0.73        50

    accuracy                           0.75       500
   macro avg       0.84      0.75      0.76       500
weighted avg       0.84      0.75      0.76       500

Generation time: 1.0171 seconds
Coverage=0.484
Avg set size=0.52
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.528    | 0.66     | 0.66    | 1.28     | 0.54    | 0.19    | 0.11    | 0      
1     | 50   | 0.552    | 0.76     | 0.76    | 1.12     | 0.56    | 0.17    | 0.10    | 1      
2     | 50   | 0.478    | 0.64     | 0.64    | 1.06     | 0.48    | 0.20    | 0.10    | 2      
3     | 50   | 0.339    | 0.38     | 0.34    | 2.00     | 0.40    | 0.23    | 0.16    | 3      
4     | 50   | 0.307    | 0.24     | 0.18    | 1.78     | 0.36    | 0.24    | 0.15    | 4      
5     | 50   | 0.357    | 0.30     | 0.20    | 1.36     | 0.39    | 0.27    | 0.12    | 5      
6     | 50   | 0.366    | 0.50     | 0.44    | 1.70     | 0.42    | 0.18    | 0.13    | 6      
7     | 50   | 0.299    | 0.24     | 0.18    | 2.06     | 0.34    | 0.26    | 0.12    | 7      
8     | 50   | 0.524    | 0.78     | 0.78    | 1.02     | 0.53    | 0.17    | 0.10    | 8      
9     | 50   | 0.504    | 0.66     | 0.66    | 1.22     | 0.52    | 0.16    | 0.11    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.766
Confusion matrix (rows=true, cols=pred):
 [[44  0  2  0  0  4  0  0  0  0]
 [ 0 48  0  0  0  0  0  0  2  0]
 [ 0  0 46  0  0  0  0  0  4  0]
 [ 1  0  8 26  0  4  0  0 11  0]
 [ 2  0  1  0 29  0  0  0  5 13]
 [ 0  0  2  1  0 34  0  0 13  0]
 [ 0  0 11  0  0  1 35  0  3  0]
 [ 0  0  2  0  0  5  0 29  6  8]
 [ 0  0  0  0  0  2  0  0 47  1]
 [ 0  0  1  0  0  1  0  0  3 45]]
Classification report:
               precision    recall  f1-score   support

           0       0.94      0.88      0.91        50
           1       1.00      0.96      0.98        50
           2       0.63      0.92      0.75        50
           3       0.96      0.52      0.68        50
           4       1.00      0.58      0.73        50
           5       0.67      0.68      0.67        50
           6       1.00      0.70      0.82        50
           7       1.00      0.58      0.73        50
           8       0.50      0.94      0.65        50
           9       0.67      0.90      0.77        50

    accuracy                           0.77       500
   macro avg       0.84      0.77      0.77       500
weighted avg       0.84      0.77      0.77       500

Generation time: 1.0364 seconds
Coverage=0.518
Avg set size=0.54
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.510    | 0.70     | 0.70    | 1.16     | 0.52    | 0.20    | 0.11    | 0      
1     | 50   | 0.695    | 0.84     | 0.84    | 1.04     | 0.70    | 0.12    | 0.07    | 1      
2     | 50   | 0.442    | 0.58     | 0.58    | 1.08     | 0.45    | 0.21    | 0.10    | 2      
3     | 50   | 0.319    | 0.38     | 0.30    | 2.18     | 0.40    | 0.23    | 0.16    | 3      
4     | 50   | 0.350    | 0.32     | 0.28    | 1.70     | 0.39    | 0.25    | 0.13    | 4      
5     | 50   | 0.365    | 0.34     | 0.30    | 1.38     | 0.39    | 0.27    | 0.11    | 5      
6     | 50   | 0.413    | 0.52     | 0.52    | 1.68     | 0.45    | 0.18    | 0.12    | 6      
7     | 50   | 0.297    | 0.24     | 0.18    | 2.20     | 0.34    | 0.26    | 0.12    | 7      
8     | 50   | 0.552    | 0.84     | 0.84    | 1.06     | 0.56    | 0.16    | 0.09    | 8      
9     | 50   | 0.514    | 0.66     | 0.64    | 1.42     | 0.54    | 0.17    | 0.11    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.682
Confusion matrix (rows=true, cols=pred):
 [[42  0  0  0  0  4  0  0  4  0]
 [ 0 48  0  0  0  0  0  0  2  0]
 [ 0  0 34  0  0  3  0  0 13  0]
 [ 0  0  1 22  0  5  0  0 22  0]
 [ 1  0  0  0 28  0  0  0 11 10]
 [ 0  0  3  0  0 29  0  0 18  0]
 [ 1  0  6  0  0  3 34  0  6  0]
 [ 0  0  2  0  0  7  0 17 12 12]
 [ 0  0  1  0  0  0  0  0 48  1]
 [ 0  0  3  0  0  2  0  0  6 39]]
Classification report:
               precision    recall  f1-score   support

           0       0.95      0.84      0.89        50
           1       1.00      0.96      0.98        50
           2       0.68      0.68      0.68        50
           3       1.00      0.44      0.61        50
           4       1.00      0.56      0.72        50
           5       0.55      0.58      0.56        50
           6       1.00      0.68      0.81        50
           7       1.00      0.34      0.51        50
           8       0.34      0.96      0.50        50
           9       0.63      0.78      0.70        50

    accuracy                           0.68       500
   macro avg       0.81      0.68      0.70       500
weighted avg       0.81      0.68      0.70       500

Generation time: 0.9583 seconds
Coverage=0.374
Avg set size=0.39
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.389    | 0.36     | 0.36    | 1.36     | 0.40    | 0.22    | 0.15    | 0      
1     | 50   | 0.739    | 0.88     | 0.86    | 1.04     | 0.75    | 0.09    | 0.06    | 1      
2     | 50   | 0.316    | 0.14     | 0.14    | 1.40     | 0.34    | 0.24    | 0.13    | 2      
3     | 50   | 0.266    | 0.22     | 0.16    | 2.22     | 0.35    | 0.23    | 0.17    | 3      
4     | 50   | 0.327    | 0.32     | 0.30    | 2.12     | 0.38    | 0.25    | 0.14    | 4      
5     | 50   | 0.328    | 0.10     | 0.10    | 1.48     | 0.35    | 0.28    | 0.11    | 5      
6     | 50   | 0.327    | 0.32     | 0.32    | 1.92     | 0.37    | 0.19    | 0.15    | 6      
7     | 50   | 0.224    | 0.12     | 0.08    | 3.26     | 0.32    | 0.25    | 0.13    | 7      
8     | 50   | 0.540    | 0.80     | 0.80    | 1.04     | 0.54    | 0.15    | 0.10    | 8      
9     | 50   | 0.466    | 0.62     | 0.62    | 1.66     | 0.50    | 0.17    | 0.11    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.3466 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.089    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.050    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.043    | 0.00     | 0.00    | 7.44     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.910
Confusion matrix (rows=true, cols=pred):
 [[39  0  2  0  0  4  1  0  4  0]
 [ 0 48  0  0  0  0  0  0  2  0]
 [ 0  0 46  2  0  1  0  0  1  0]
 [ 0  0  1 48  0  0  0  1  0  0]
 [ 0  0  1  0 44  0  0  0  0  5]
 [ 0  1  0  2  2 41  0  0  4  0]
 [ 0  0  0  1  0  0 48  0  1  0]
 [ 0  0  0  0  0  0  0 49  0  1]
 [ 0  1  2  3  0  0  0  0 43  1]
 [ 0  0  0  0  0  0  0  0  1 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.78      0.88        50
           1       0.96      0.96      0.96        50
           2       0.88      0.92      0.90        50
           3       0.86      0.96      0.91        50
           4       0.96      0.88      0.92        50
           5       0.89      0.82      0.85        50
           6       0.98      0.96      0.97        50
           7       0.98      0.98      0.98        50
           8       0.77      0.86      0.81        50
           9       0.88      0.98      0.92        50

    accuracy                           0.91       500
   macro avg       0.92      0.91      0.91       500
weighted avg       0.92      0.91      0.91       500

Generation time: 1.3265 seconds
Coverage=0.820
Avg set size=0.86
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.682    | 0.74     | 0.68    | 1.66     | 0.73    | 0.13    | 0.05    | 0      
1     | 50   | 0.550    | 0.76     | 0.76    | 1.06     | 0.55    | 0.18    | 0.10    | 1      
2     | 50   | 0.810    | 0.94     | 0.90    | 1.14     | 0.82    | 0.08    | 0.04    | 2      
3     | 50   | 0.807    | 0.92     | 0.92    | 1.04     | 0.81    | 0.09    | 0.04    | 3      
4     | 50   | 0.544    | 0.78     | 0.70    | 1.14     | 0.57    | 0.20    | 0.08    | 4      
5     | 50   | 0.579    | 0.88     | 0.74    | 1.40     | 0.62    | 0.20    | 0.09    | 5      
6     | 50   | 0.777    | 0.90     | 0.90    | 1.14     | 0.79    | 0.11    | 0.05    | 6      
7     | 50   | 0.816    | 0.94     | 0.92    | 1.02     | 0.82    | 0.12    | 0.03    | 7      
8     | 50   | 0.602    | 0.84     | 0.78    | 1.16     | 0.62    | 0.18    | 0.08    | 8      
9     | 50   | 0.671    | 0.90     | 0.90    | 1.02     | 0.68    | 0.15    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=3.60332e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.1456 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 7.96     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=7.83564e-09): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.786
Confusion matrix (rows=true, cols=pred):
 [[29  0  4  1  1  8  2  1  4  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 1  0 41  0  1  1  3  1  2  0]
 [ 0  0  4 42  0  0  0  1  2  1]
 [ 0  0  0  0 33  1  2  0  3 11]
 [ 0  1  1  3  2 36  1  2  2  2]
 [ 0  1  0  0  0  4 44  0  1  0]
 [ 0  0  0  0  0  0  0 40  2  8]
 [ 0  3  2  4  0  4  0  0 37  0]
 [ 0  1  0  0  4  0  0  4  0 41]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.58      0.72        50
           1       0.89      1.00      0.94        50
           2       0.79      0.82      0.80        50
           3       0.84      0.84      0.84        50
           4       0.80      0.66      0.73        50
           5       0.67      0.72      0.69        50
           6       0.85      0.88      0.86        50
           7       0.82      0.80      0.81        50
           8       0.70      0.74      0.72        50
           9       0.65      0.82      0.73        50

    accuracy                           0.79       500
   macro avg       0.80      0.79      0.78       500
weighted avg       0.80      0.79      0.78       500

Generation time: 1.1060 seconds
Coverage=0.754
Avg set size=0.90
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.546    | 0.92     | 0.58    | 2.82     | 0.76    | 0.12    | 0.05    | 0      
1     | 50   | 0.666    | 0.90     | 0.90    | 1.00     | 0.67    | 0.12    | 0.08    | 1      
2     | 50   | 0.741    | 0.94     | 0.82    | 1.54     | 0.80    | 0.10    | 0.04    | 2      
3     | 50   | 0.759    | 0.90     | 0.82    | 1.38     | 0.81    | 0.11    | 0.05    | 3      
4     | 50   | 0.494    | 0.78     | 0.56    | 1.54     | 0.57    | 0.21    | 0.09    | 4      
5     | 50   | 0.522    | 0.84     | 0.68    | 1.74     | 0.62    | 0.20    | 0.09    | 5      
6     | 50   | 0.771    | 0.96     | 0.86    | 1.34     | 0.81    | 0.09    | 0.04    | 6      
7     | 50   | 0.741    | 0.94     | 0.80    | 1.42     | 0.81    | 0.13    | 0.03    | 7      
8     | 50   | 0.560    | 0.86     | 0.70    | 1.32     | 0.62    | 0.21    | 0.08    | 8      
9     | 50   | 0.659    | 0.98     | 0.82    | 1.18     | 0.70    | 0.18    | 0.06    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.516
Confusion matrix (rows=true, cols=pred):
 [[36  0  0  0  0  9  0  0  5  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 41  0  0  0  0  0  9  0]
 [ 0  0  1 26  0  2  0  0 21  0]
 [ 0  0  0  0  2  0  0  0 31 17]
 [ 0  0  0  0  0 17  0  0 33  0]
 [ 0  0  6  0  0  1 28  0 15  0]
 [ 0  0  0  0  0  0  0 19 10 21]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 11 39]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.72      0.84        50
           1       0.00      0.00      0.00        50
           2       0.85      0.82      0.84        50
           3       1.00      0.52      0.68        50
           4       1.00      0.04      0.08        50
           5       0.59      0.34      0.43        50
           6       1.00      0.56      0.72        50
           7       1.00      0.38      0.55        50
           8       0.21      1.00      0.35        50
           9       0.51      0.78      0.61        50

    accuracy                           0.52       500
   macro avg       0.72      0.52      0.51       500
weighted avg       0.72      0.52      0.51       500

Generation time: 0.9867 seconds
Coverage=0.232
Avg set size=0.24
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.409    | 0.48     | 0.46    | 1.84     | 0.46    | 0.21    | 0.13    | 0      
1     | 50   | 0.131    | 0.00     | 0.00    | 3.20     | 0.33    | 0.15    | 0.13    | 8      
2     | 50   | 0.437    | 0.52     | 0.52    | 1.22     | 0.45    | 0.21    | 0.11    | 2      
3     | 50   | 0.302    | 0.22     | 0.20    | 1.72     | 0.36    | 0.23    | 0.17    | 3      
4     | 50   | 0.162    | 0.00     | 0.00    | 2.94     | 0.27    | 0.21    | 0.16    | 8      
5     | 50   | 0.297    | 0.08     | 0.02    | 1.70     | 0.36    | 0.27    | 0.11    | 8      
6     | 50   | 0.275    | 0.12     | 0.10    | 1.86     | 0.34    | 0.19    | 0.15    | 6      
7     | 50   | 0.283    | 0.10     | 0.10    | 1.98     | 0.33    | 0.25    | 0.15    | 9      
8     | 50   | 0.468    | 0.76     | 0.76    | 1.00     | 0.47    | 0.17    | 0.10    | 8      
9     | 50   | 0.344    | 0.16     | 0.16    | 1.22     | 0.35    | 0.22    | 0.12    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9996 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.069    | 0.00     | 0.00    | 5.54     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.038    | 0.00     | 0.00    | 8.76     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.157    | 0.00     | 0.00    | 2.42     | 0.34    | 0.16    | 0.14    | 8      
3     | 50   | 0.117    | 0.00     | 0.00    | 3.64     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.02     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.185    | 0.00     | 0.00    | 2.00     | 0.35    | 0.19    | 0.11    | 8      
6     | 50   | 0.071    | 0.00     | 0.00    | 5.56     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 5.92     | 0.33    | 0.15    | 0.14    | 8      
8     | 50   | 0.371    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.142    | 0.00     | 0.00    | 2.58     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=12 | std=1.0 ---

== PCA [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.822
Confusion matrix (rows=true, cols=pred):
 [[36  0  3  0  0 11  0  0  0  0]
 [ 0 46  0  0  0  0  0  0  4  0]
 [ 0  0 32  2  0  9  1  0  3  3]
 [ 0  0  5 32  0  6  0  1  6  0]
 [ 0  0  0  0 43  2  0  1  2  2]
 [ 0  0  0  1  2 45  0  0  2  0]
 [ 0  0  0  0  1  0 47  0  2  0]
 [ 0  0  0  0  1  0  0 42  0  7]
 [ 0  0  2  0  0  1  1  0 44  2]
 [ 0  0  0  0  0  1  0  0  5 44]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.72      0.84        50
           1       1.00      0.92      0.96        50
           2       0.76      0.64      0.70        50
           3       0.91      0.64      0.75        50
           4       0.91      0.86      0.89        50
           5       0.60      0.90      0.72        50
           6       0.96      0.94      0.95        50
           7       0.95      0.84      0.89        50
           8       0.65      0.88      0.75        50
           9       0.76      0.88      0.81        50

    accuracy                           0.82       500
   macro avg       0.85      0.82      0.83       500
weighted avg       0.85      0.82      0.83       500

Generation time: 0.6979 seconds
Coverage=0.690
Avg set size=0.75
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.459    | 0.68     | 0.56    | 1.36     | 0.52    | 0.22    | 0.11    | 0      
1     | 50   | 0.516    | 0.78     | 0.76    | 1.10     | 0.53    | 0.22    | 0.10    | 1      
2     | 50   | 0.435    | 0.60     | 0.56    | 2.08     | 0.52    | 0.17    | 0.10    | 2      
3     | 50   | 0.416    | 0.62     | 0.50    | 1.60     | 0.47    | 0.21    | 0.11    | 3      
4     | 50   | 0.597    | 0.74     | 0.70    | 1.16     | 0.61    | 0.16    | 0.10    | 4      
5     | 50   | 0.524    | 0.74     | 0.70    | 1.12     | 0.54    | 0.20    | 0.09    | 5      
6     | 50   | 0.546    | 0.74     | 0.72    | 1.08     | 0.56    | 0.16    | 0.10    | 6      
7     | 50   | 0.578    | 1.00     | 0.82    | 1.16     | 0.60    | 0.24    | 0.06    | 7      
8     | 50   | 0.537    | 0.76     | 0.76    | 1.14     | 0.55    | 0.20    | 0.09    | 8      
9     | 50   | 0.580    | 0.84     | 0.82    | 1.14     | 0.59    | 0.17    | 0.09    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.848
Confusion matrix (rows=true, cols=pred):
 [[49  0  0  0  0  1  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  3 33  0  1  0  0 13  0]
 [ 0  0  0  0 29  0  0  0  6 15]
 [ 0  0  0  0  0 40  0  0 10  0]
 [ 0  0  3  0  0  0 44  0  3  0]
 [ 0  0  0  0  0  2  0 31  9  8]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  1 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.98      0.99        50
           1       1.00      1.00      1.00        50
           2       0.89      0.98      0.93        50
           3       1.00      0.66      0.80        50
           4       1.00      0.58      0.73        50
           5       0.91      0.80      0.85        50
           6       1.00      0.88      0.94        50
           7       1.00      0.62      0.77        50
           8       0.54      1.00      0.70        50
           9       0.68      0.98      0.80        50

    accuracy                           0.85       500
   macro avg       0.90      0.85      0.85       500
weighted avg       0.90      0.85      0.85       500

Generation time: 1.6744 seconds
Coverage=0.512
Avg set size=0.51
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.523    | 0.70     | 0.70    | 1.02     | 0.52    | 0.20    | 0.12    | 0      
1     | 50   | 0.564    | 0.96     | 0.96    | 1.00     | 0.56    | 0.19    | 0.08    | 1      
2     | 50   | 0.497    | 0.82     | 0.82    | 1.02     | 0.50    | 0.19    | 0.09    | 2      
3     | 50   | 0.329    | 0.16     | 0.16    | 1.44     | 0.35    | 0.25    | 0.18    | 3      
4     | 50   | 0.288    | 0.06     | 0.06    | 1.68     | 0.31    | 0.26    | 0.17    | 4      
5     | 50   | 0.373    | 0.20     | 0.20    | 1.20     | 0.38    | 0.30    | 0.09    | 5      
6     | 50   | 0.392    | 0.32     | 0.32    | 1.16     | 0.40    | 0.19    | 0.14    | 6      
7     | 50   | 0.290    | 0.12     | 0.10    | 1.78     | 0.31    | 0.26    | 0.13    | 7      
8     | 50   | 0.625    | 1.00     | 1.00    | 1.00     | 0.62    | 0.12    | 0.08    | 8      
9     | 50   | 0.490    | 0.80     | 0.80    | 1.02     | 0.49    | 0.16    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.858
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  2 31  0  4  0  0 13  0]
 [ 0  0  1  0 35  0  0  0  5  9]
 [ 0  0  0  0  0 42  0  0  8  0]
 [ 0  0  4  0  0  1 43  0  2  0]
 [ 0  0  0  0  0  1  0 31 10  8]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  2 48]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       0.88      0.98      0.92        50
           3       1.00      0.62      0.77        50
           4       1.00      0.70      0.82        50
           5       0.88      0.84      0.86        50
           6       1.00      0.86      0.92        50
           7       1.00      0.62      0.77        50
           8       0.55      1.00      0.71        50
           9       0.74      0.96      0.83        50

    accuracy                           0.86       500
   macro avg       0.90      0.86      0.86       500
weighted avg       0.90      0.86      0.86       500

Generation time: 1.0518 seconds
Coverage=0.540
Avg set size=0.54
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.516    | 0.72     | 0.72    | 1.00     | 0.52    | 0.20    | 0.11    | 0      
1     | 50   | 0.714    | 1.00     | 1.00    | 1.00     | 0.71    | 0.12    | 0.06    | 1      
2     | 50   | 0.453    | 0.60     | 0.60    | 1.02     | 0.45    | 0.21    | 0.11    | 2      
3     | 50   | 0.335    | 0.18     | 0.18    | 1.54     | 0.36    | 0.25    | 0.18    | 3      
4     | 50   | 0.322    | 0.24     | 0.24    | 1.50     | 0.35    | 0.26    | 0.15    | 4      
5     | 50   | 0.383    | 0.22     | 0.22    | 1.16     | 0.39    | 0.29    | 0.09    | 5      
6     | 50   | 0.422    | 0.48     | 0.48    | 1.24     | 0.43    | 0.17    | 0.14    | 6      
7     | 50   | 0.293    | 0.16     | 0.16    | 1.90     | 0.32    | 0.26    | 0.13    | 7      
8     | 50   | 0.663    | 0.98     | 0.98    | 1.00     | 0.66    | 0.11    | 0.07    | 8      
9     | 50   | 0.518    | 0.82     | 0.82    | 1.06     | 0.52    | 0.16    | 0.12    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.704
Confusion matrix (rows=true, cols=pred):
 [[41  0  0  0  0  9  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 34  0  0  0  0  0 16  0]
 [ 0  0  2 20  0  2  0  0 26  0]
 [ 0  0  1  0 24  0  0  0 18  7]
 [ 0  0  0  0  0 37  0  0 13  0]
 [ 0  0  4  0  0  2 35  0  9  0]
 [ 0  0  0  0  0  2  0 15 21 12]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  4 46]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.82      0.90        50
           1       1.00      1.00      1.00        50
           2       0.83      0.68      0.75        50
           3       1.00      0.40      0.57        50
           4       1.00      0.48      0.65        50
           5       0.71      0.74      0.73        50
           6       1.00      0.70      0.82        50
           7       1.00      0.30      0.46        50
           8       0.32      1.00      0.48        50
           9       0.71      0.92      0.80        50

    accuracy                           0.70       500
   macro avg       0.86      0.70      0.72       500
weighted avg       0.86      0.70      0.72       500

Generation time: 1.0691 seconds
Coverage=0.328
Avg set size=0.33
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.369    | 0.26     | 0.26    | 1.30     | 0.38    | 0.23    | 0.16    | 0      
1     | 50   | 0.775    | 1.00     | 1.00    | 1.00     | 0.77    | 0.10    | 0.05    | 1      
2     | 50   | 0.304    | 0.02     | 0.02    | 1.34     | 0.32    | 0.25    | 0.14    | 2      
3     | 50   | 0.263    | 0.04     | 0.04    | 2.10     | 0.32    | 0.25    | 0.19    | 8      
4     | 50   | 0.282    | 0.14     | 0.14    | 1.94     | 0.33    | 0.24    | 0.16    | 4      
5     | 50   | 0.340    | 0.04     | 0.04    | 1.26     | 0.35    | 0.29    | 0.10    | 5      
6     | 50   | 0.303    | 0.16     | 0.16    | 1.70     | 0.33    | 0.19    | 0.16    | 6      
7     | 50   | 0.209    | 0.06     | 0.06    | 3.08     | 0.28    | 0.23    | 0.14    | 8      
8     | 50   | 0.627    | 0.98     | 0.98    | 1.00     | 0.63    | 0.12    | 0.08    | 8      
9     | 50   | 0.465    | 0.58     | 0.58    | 1.08     | 0.47    | 0.17    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9368 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 7.98     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.988
Confusion matrix (rows=true, cols=pred):
 [[49  0  0  0  0  1  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 50  0  0  0  0  0  0]
 [ 0  0  0  0 47  0  0  0  0  3]
 [ 0  0  0  0  0 49  0  0  1  0]
 [ 0  0  0  0  0  0 50  0  0  0]
 [ 0  0  0  0  0  0  0 49  0  1]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.98      0.99        50
           1       1.00      1.00      1.00        50
           2       1.00      1.00      1.00        50
           3       1.00      1.00      1.00        50
           4       1.00      0.94      0.97        50
           5       0.98      0.98      0.98        50
           6       1.00      1.00      1.00        50
           7       1.00      0.98      0.99        50
           8       0.98      1.00      0.99        50
           9       0.93      1.00      0.96        50

    accuracy                           0.99       500
   macro avg       0.99      0.99      0.99       500
weighted avg       0.99      0.99      0.99       500

Generation time: 0.9446 seconds
Coverage=0.974
Avg set size=0.98
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.841    | 0.94     | 0.94    | 1.06     | 0.84    | 0.08    | 0.04    | 0      
1     | 50   | 0.599    | 1.00     | 1.00    | 1.00     | 0.60    | 0.16    | 0.08    | 1      
2     | 50   | 0.921    | 1.00     | 1.00    | 1.00     | 0.92    | 0.03    | 0.02    | 2      
3     | 50   | 0.804    | 0.98     | 0.98    | 1.00     | 0.80    | 0.09    | 0.05    | 3      
4     | 50   | 0.609    | 0.94     | 0.90    | 1.06     | 0.61    | 0.20    | 0.07    | 4      
5     | 50   | 0.639    | 0.96     | 0.96    | 1.02     | 0.64    | 0.21    | 0.06    | 5      
6     | 50   | 0.867    | 0.98     | 0.98    | 1.00     | 0.87    | 0.06    | 0.03    | 6      
7     | 50   | 0.821    | 1.04     | 1.00    | 1.02     | 0.82    | 0.13    | 0.02    | 7      
8     | 50   | 0.740    | 0.98     | 0.98    | 1.00     | 0.74    | 0.11    | 0.06    | 8      
9     | 50   | 0.684    | 1.00     | 1.00    | 1.00     | 0.68    | 0.14    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=4.1806e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.1414 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=1.095e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.982
Confusion matrix (rows=true, cols=pred):
 [[49  0  0  0  0  1  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 50  0  0  0  0  0  0]
 [ 0  0  0  0 48  0  0  0  0  2]
 [ 0  0  0  2  0 47  0  0  1  0]
 [ 0  0  0  0  0  0 50  0  0  0]
 [ 0  0  0  0  0  0  0 48  0  2]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  1  0 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.98      0.99        50
           1       1.00      1.00      1.00        50
           2       1.00      1.00      1.00        50
           3       0.96      1.00      0.98        50
           4       1.00      0.96      0.98        50
           5       0.98      0.94      0.96        50
           6       1.00      1.00      1.00        50
           7       0.98      0.96      0.97        50
           8       0.98      1.00      0.99        50
           9       0.92      0.98      0.95        50

    accuracy                           0.98       500
   macro avg       0.98      0.98      0.98       500
weighted avg       0.98      0.98      0.98       500

Generation time: 1.1514 seconds
Coverage=0.972
Avg set size=0.99
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.861    | 1.00     | 0.98    | 1.02     | 0.86    | 0.09    | 0.02    | 0      
1     | 50   | 0.808    | 1.00     | 1.00    | 1.00     | 0.81    | 0.08    | 0.04    | 1      
2     | 50   | 0.949    | 1.00     | 1.00    | 1.00     | 0.95    | 0.03    | 0.01    | 2      
3     | 50   | 0.927    | 0.98     | 0.98    | 1.00     | 0.93    | 0.04    | 0.02    | 3      
4     | 50   | 0.761    | 1.00     | 0.96    | 1.04     | 0.77    | 0.15    | 0.03    | 4      
5     | 50   | 0.740    | 0.96     | 0.90    | 1.06     | 0.75    | 0.15    | 0.05    | 5      
6     | 50   | 0.955    | 1.00     | 1.00    | 1.00     | 0.95    | 0.02    | 0.01    | 6      
7     | 50   | 0.869    | 1.00     | 0.96    | 1.04     | 0.87    | 0.10    | 0.01    | 7      
8     | 50   | 0.801    | 0.96     | 0.96    | 1.00     | 0.80    | 0.09    | 0.04    | 8      
9     | 50   | 0.780    | 0.98     | 0.98    | 1.02     | 0.78    | 0.12    | 0.05    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.550
Confusion matrix (rows=true, cols=pred):
 [[44  0  0  0  0  3  0  0  3  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 48  0  0  0  0  0  2  0]
 [ 0  0  0 27  0  0  0  0 23  0]
 [ 0  0  0  0  0  0  0  0 35 15]
 [ 0  0  0  0  0 10  0  0 40  0]
 [ 0  0  3  0  0  0 37  0 10  0]
 [ 0  0  0  0  0  0  0 17  3 30]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  8 42]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.88      0.94        50
           1       0.00      0.00      0.00        50
           2       0.94      0.96      0.95        50
           3       1.00      0.54      0.70        50
           4       0.00      0.00      0.00        50
           5       0.77      0.20      0.32        50
           6       1.00      0.74      0.85        50
           7       1.00      0.34      0.51        50
           8       0.22      1.00      0.36        50
           9       0.48      0.84      0.61        50

    accuracy                           0.55       500
   macro avg       0.64      0.55      0.52       500
weighted avg       0.64      0.55      0.52       500

Generation time: 1.0604 seconds
Coverage=0.224
Avg set size=0.22
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.422    | 0.42     | 0.42    | 1.16     | 0.43    | 0.22    | 0.15    | 0      
1     | 50   | 0.132    | 0.00     | 0.00    | 2.34     | 0.33    | 0.14    | 0.12    | 8      
2     | 50   | 0.475    | 0.72     | 0.72    | 1.04     | 0.48    | 0.22    | 0.09    | 2      
3     | 50   | 0.295    | 0.04     | 0.04    | 1.50     | 0.32    | 0.26    | 0.18    | 3      
4     | 50   | 0.164    | 0.00     | 0.00    | 2.96     | 0.26    | 0.22    | 0.16    | 8      
5     | 50   | 0.299    | 0.00     | 0.00    | 1.80     | 0.35    | 0.29    | 0.09    | 8      
6     | 50   | 0.287    | 0.08     | 0.08    | 1.38     | 0.30    | 0.21    | 0.17    | 6      
7     | 50   | 0.277    | 0.02     | 0.02    | 1.86     | 0.31    | 0.26    | 0.16    | 9      
8     | 50   | 0.498    | 0.92     | 0.92    | 1.00     | 0.50    | 0.15    | 0.10    | 8      
9     | 50   | 0.333    | 0.04     | 0.04    | 1.16     | 0.34    | 0.24    | 0.12    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0587 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.069    | 0.00     | 0.00    | 5.42     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.039    | 0.00     | 0.00    | 8.80     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.162    | 0.00     | 0.00    | 2.26     | 0.34    | 0.16    | 0.15    | 8      
3     | 50   | 0.116    | 0.00     | 0.00    | 3.56     | 0.35    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.02     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.184    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.10    | 8      
6     | 50   | 0.071    | 0.00     | 0.00    | 5.72     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 6.00     | 0.32    | 0.15    | 0.14    | 8      
8     | 50   | 0.374    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.141    | 0.00     | 0.00    | 2.78     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=12 | std=1.5 ---

== PCA [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.692
Confusion matrix (rows=true, cols=pred):
 [[28  0  5  1  0 13  1  1  0  1]
 [ 0 39  0  3  0  1  0  0  7  0]
 [ 2  0 29  2  0 10  2  1  3  1]
 [ 1  0  9 27  0  6  0  1  5  1]
 [ 0  0  0  0 37  4  0  2  4  3]
 [ 1  0  2  5  3 34  0  0  5  0]
 [ 0  0  3  0  2  4 40  0  1  0]
 [ 0  0  1  0  1  0  0 40  0  8]
 [ 0  0  6  1  1  4  2  0 33  3]
 [ 0  0  1  0  0  2  0  3  5 39]]
Classification report:
               precision    recall  f1-score   support

           0       0.88      0.56      0.68        50
           1       1.00      0.78      0.88        50
           2       0.52      0.58      0.55        50
           3       0.69      0.54      0.61        50
           4       0.84      0.74      0.79        50
           5       0.44      0.68      0.53        50
           6       0.89      0.80      0.84        50
           7       0.83      0.80      0.82        50
           8       0.52      0.66      0.58        50
           9       0.70      0.78      0.74        50

    accuracy                           0.69       500
   macro avg       0.73      0.69      0.70       500
weighted avg       0.73      0.69      0.70       500

Generation time: 1.0319 seconds
Coverage=0.592
Avg set size=0.76
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.408    | 0.70     | 0.50    | 1.80     | 0.54    | 0.22    | 0.10    | 0      
1     | 50   | 0.520    | 0.70     | 0.62    | 1.34     | 0.56    | 0.20    | 0.10    | 1      
2     | 50   | 0.449    | 0.72     | 0.52    | 2.48     | 0.59    | 0.17    | 0.09    | 2      
3     | 50   | 0.430    | 0.74     | 0.48    | 1.94     | 0.54    | 0.20    | 0.10    | 3      
4     | 50   | 0.531    | 0.76     | 0.60    | 1.38     | 0.62    | 0.16    | 0.09    | 4      
5     | 50   | 0.496    | 0.72     | 0.56    | 1.50     | 0.56    | 0.20    | 0.09    | 5      
6     | 50   | 0.548    | 0.70     | 0.66    | 1.32     | 0.58    | 0.18    | 0.09    | 6      
7     | 50   | 0.584    | 0.94     | 0.76    | 1.22     | 0.63    | 0.21    | 0.07    | 7      
8     | 50   | 0.449    | 0.78     | 0.54    | 1.48     | 0.53    | 0.24    | 0.10    | 8      
9     | 50   | 0.535    | 0.80     | 0.68    | 1.30     | 0.59    | 0.19    | 0.08    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.784
Confusion matrix (rows=true, cols=pred):
 [[46  0  0  0  0  4  0  0  0  0]
 [ 0 48  0  0  0  0  0  0  2  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  5 30  0  4  0  0 11  0]
 [ 0  0  1  0 26  0  0  0 10 13]
 [ 0  0  1  0  0 36  0  0 13  0]
 [ 0  0  8  0  0  1 34  0  7  0]
 [ 0  0  3  0  0  5  0 27  9  6]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  4 46]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.92      0.96        50
           1       1.00      0.96      0.98        50
           2       0.73      0.98      0.84        50
           3       1.00      0.60      0.75        50
           4       1.00      0.52      0.68        50
           5       0.72      0.72      0.72        50
           6       1.00      0.68      0.81        50
           7       1.00      0.54      0.70        50
           8       0.47      1.00      0.64        50
           9       0.71      0.92      0.80        50

    accuracy                           0.78       500
   macro avg       0.86      0.78      0.79       500
weighted avg       0.86      0.78      0.79       500

Generation time: 1.0471 seconds
Coverage=0.486
Avg set size=0.50
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.507    | 0.60     | 0.60    | 1.12     | 0.51    | 0.20    | 0.13    | 0      
1     | 50   | 0.537    | 0.78     | 0.74    | 1.04     | 0.54    | 0.20    | 0.09    | 1      
2     | 50   | 0.491    | 0.72     | 0.72    | 1.02     | 0.49    | 0.19    | 0.10    | 2      
3     | 50   | 0.325    | 0.26     | 0.20    | 1.68     | 0.38    | 0.24    | 0.17    | 3      
4     | 50   | 0.276    | 0.12     | 0.12    | 1.94     | 0.32    | 0.25    | 0.15    | 4      
5     | 50   | 0.361    | 0.32     | 0.28    | 1.32     | 0.38    | 0.27    | 0.11    | 5      
6     | 50   | 0.386    | 0.34     | 0.34    | 1.56     | 0.41    | 0.19    | 0.13    | 6      
7     | 50   | 0.286    | 0.22     | 0.18    | 2.52     | 0.34    | 0.26    | 0.12    | 7      
8     | 50   | 0.610    | 0.96     | 0.96    | 1.00     | 0.61    | 0.13    | 0.09    | 8      
9     | 50   | 0.480    | 0.72     | 0.72    | 1.18     | 0.49    | 0.17    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.782
Confusion matrix (rows=true, cols=pred):
 [[44  0  0  0  0  5  0  0  1  0]
 [ 0 47  0  0  0  0  0  0  3  0]
 [ 0  0 47  0  0  0  0  0  3  0]
 [ 0  0  4 30  0  4  0  0 12  0]
 [ 1  0  1  0 31  0  0  0  8  9]
 [ 0  0  3  0  0 37  0  0 10  0]
 [ 2  0  4  0  0  2 37  0  5  0]
 [ 0  0  2  0  0  4  0 24 10 10]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  1  1  0  0  4 44]]
Classification report:
               precision    recall  f1-score   support

           0       0.94      0.88      0.91        50
           1       1.00      0.94      0.97        50
           2       0.77      0.94      0.85        50
           3       1.00      0.60      0.75        50
           4       0.97      0.62      0.76        50
           5       0.70      0.74      0.72        50
           6       1.00      0.74      0.85        50
           7       1.00      0.48      0.65        50
           8       0.47      1.00      0.64        50
           9       0.70      0.88      0.78        50

    accuracy                           0.78       500
   macro avg       0.85      0.78      0.79       500
weighted avg       0.85      0.78      0.79       500

Generation time: 1.0573 seconds
Coverage=0.528
Avg set size=0.54
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.505    | 0.64     | 0.64    | 1.16     | 0.51    | 0.20    | 0.12    | 0      
1     | 50   | 0.674    | 0.96     | 0.92    | 1.06     | 0.68    | 0.14    | 0.07    | 1      
2     | 50   | 0.451    | 0.54     | 0.54    | 1.08     | 0.46    | 0.21    | 0.11    | 2      
3     | 50   | 0.334    | 0.28     | 0.28    | 1.78     | 0.39    | 0.23    | 0.16    | 3      
4     | 50   | 0.309    | 0.32     | 0.28    | 1.90     | 0.36    | 0.25    | 0.14    | 4      
5     | 50   | 0.368    | 0.26     | 0.24    | 1.30     | 0.38    | 0.26    | 0.11    | 5      
6     | 50   | 0.423    | 0.48     | 0.48    | 1.62     | 0.45    | 0.18    | 0.13    | 6      
7     | 50   | 0.291    | 0.28     | 0.26    | 2.66     | 0.34    | 0.26    | 0.12    | 7      
8     | 50   | 0.652    | 0.92     | 0.92    | 1.00     | 0.65    | 0.13    | 0.08    | 8      
9     | 50   | 0.502    | 0.72     | 0.72    | 1.24     | 0.51    | 0.17    | 0.11    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.656
Confusion matrix (rows=true, cols=pred):
 [[36  0  0  0  0  7  0  0  7  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 33  0  0  0  0  0 17  0]
 [ 0  0  4 21  0  7  0  0 18  0]
 [ 0  0  1  0 23  1  0  0 17  8]
 [ 0  0  2  0  0 34  0  0 14  0]
 [ 1  0  4  0  0  4 29  0 12  0]
 [ 0  0  2  0  0  5  0 15 17 11]
 [ 0  0  0  0  0  1  0  0 49  0]
 [ 0  0  0  0  0  0  0  0 11 39]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.72      0.83        50
           1       1.00      0.98      0.99        50
           2       0.72      0.66      0.69        50
           3       1.00      0.42      0.59        50
           4       1.00      0.46      0.63        50
           5       0.58      0.68      0.62        50
           6       1.00      0.58      0.73        50
           7       1.00      0.30      0.46        50
           8       0.30      0.98      0.46        50
           9       0.67      0.78      0.72        50

    accuracy                           0.66       500
   macro avg       0.82      0.66      0.67       500
weighted avg       0.82      0.66      0.67       500

Generation time: 1.0664 seconds
Coverage=0.366
Avg set size=0.38
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.374    | 0.30     | 0.30    | 1.60     | 0.39    | 0.22    | 0.15    | 0      
1     | 50   | 0.719    | 0.96     | 0.94    | 1.04     | 0.72    | 0.12    | 0.06    | 1      
2     | 50   | 0.305    | 0.12     | 0.10    | 1.42     | 0.34    | 0.24    | 0.14    | 2      
3     | 50   | 0.271    | 0.16     | 0.16    | 2.30     | 0.34    | 0.24    | 0.17    | 3      
4     | 50   | 0.279    | 0.22     | 0.16    | 2.40     | 0.36    | 0.24    | 0.14    | 4      
5     | 50   | 0.329    | 0.12     | 0.12    | 1.34     | 0.35    | 0.27    | 0.11    | 5      
6     | 50   | 0.311    | 0.28     | 0.28    | 2.00     | 0.36    | 0.19    | 0.15    | 6      
7     | 50   | 0.215    | 0.14     | 0.14    | 3.74     | 0.32    | 0.24    | 0.13    | 8      
8     | 50   | 0.620    | 0.92     | 0.92    | 1.02     | 0.62    | 0.13    | 0.08    | 8      
9     | 50   | 0.444    | 0.54     | 0.54    | 1.42     | 0.47    | 0.17    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9275 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.049    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.042    | 0.00     | 0.00    | 7.72     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.954
Confusion matrix (rows=true, cols=pred):
 [[46  0  2  0  0  0  1  0  1  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  0 49  0  1  0  0  0  0]
 [ 0  0  0  0 45  0  0  0  2  3]
 [ 0  0  1  0  0 45  1  0  3  0]
 [ 0  0  0  0  0  2 48  0  0  0]
 [ 0  0  0  0  0  0  0 47  0  3]
 [ 0  0  1  0  0  0  0  0 49  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.92      0.96        50
           1       1.00      0.98      0.99        50
           2       0.92      0.98      0.95        50
           3       1.00      0.98      0.99        50
           4       1.00      0.90      0.95        50
           5       0.94      0.90      0.92        50
           6       0.96      0.96      0.96        50
           7       1.00      0.94      0.97        50
           8       0.86      0.98      0.92        50
           9       0.89      1.00      0.94        50

    accuracy                           0.95       500
   macro avg       0.96      0.95      0.95       500
weighted avg       0.96      0.95      0.95       500

Generation time: 0.9523 seconds
Coverage=0.892
Avg set size=0.91
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.771    | 0.86     | 0.82    | 1.28     | 0.79    | 0.10    | 0.05    | 0      
1     | 50   | 0.595    | 0.92     | 0.92    | 1.02     | 0.60    | 0.16    | 0.09    | 1      
2     | 50   | 0.890    | 0.92     | 0.92    | 1.02     | 0.89    | 0.05    | 0.02    | 2      
3     | 50   | 0.794    | 0.98     | 0.98    | 1.02     | 0.79    | 0.09    | 0.05    | 3      
4     | 50   | 0.596    | 0.86     | 0.80    | 1.10     | 0.61    | 0.18    | 0.08    | 4      
5     | 50   | 0.621    | 0.82     | 0.78    | 1.10     | 0.63    | 0.19    | 0.08    | 5      
6     | 50   | 0.831    | 1.00     | 0.96    | 1.06     | 0.84    | 0.08    | 0.03    | 6      
7     | 50   | 0.812    | 0.98     | 0.94    | 1.06     | 0.82    | 0.12    | 0.03    | 7      
8     | 50   | 0.691    | 0.92     | 0.92    | 1.04     | 0.69    | 0.14    | 0.07    | 8      
9     | 50   | 0.666    | 0.88     | 0.88    | 1.00     | 0.67    | 0.16    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=4.1806e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.1586 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=1.095e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.926
Confusion matrix (rows=true, cols=pred):
 [[39  0  1  1  0  7  1  0  1  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 47  1  0  0  0  0  2  0]
 [ 0  0  0 49  0  0  0  0  1  0]
 [ 0  0  0  0 43  0  0  0  0  7]
 [ 0  0  0  2  1 45  0  0  1  1]
 [ 0  0  0  0  0  0 50  0  0  0]
 [ 0  0  1  0  0  0  0 46  0  3]
 [ 0  2  0  1  0  2  0  0 45  0]
 [ 0  0  0  0  0  0  0  1  0 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.78      0.88        50
           1       0.96      1.00      0.98        50
           2       0.96      0.94      0.95        50
           3       0.91      0.98      0.94        50
           4       0.98      0.86      0.91        50
           5       0.83      0.90      0.87        50
           6       0.98      1.00      0.99        50
           7       0.98      0.92      0.95        50
           8       0.90      0.90      0.90        50
           9       0.82      0.98      0.89        50

    accuracy                           0.93       500
   macro avg       0.93      0.93      0.93       500
weighted avg       0.93      0.93      0.93       500

Generation time: 1.1366 seconds
Coverage=0.916
Avg set size=0.97
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.738    | 0.96     | 0.80    | 1.28     | 0.79    | 0.13    | 0.04    | 0      
1     | 50   | 0.778    | 0.98     | 0.98    | 1.00     | 0.78    | 0.09    | 0.05    | 1      
2     | 50   | 0.903    | 0.98     | 0.94    | 1.08     | 0.91    | 0.04    | 0.02    | 2      
3     | 50   | 0.910    | 0.98     | 0.98    | 1.04     | 0.91    | 0.05    | 0.02    | 3      
4     | 50   | 0.712    | 0.92     | 0.82    | 1.22     | 0.74    | 0.15    | 0.05    | 4      
5     | 50   | 0.704    | 0.92     | 0.86    | 1.14     | 0.74    | 0.15    | 0.05    | 5      
6     | 50   | 0.915    | 1.02     | 1.00    | 1.00     | 0.92    | 0.05    | 0.01    | 6      
7     | 50   | 0.833    | 1.00     | 0.94    | 1.10     | 0.85    | 0.10    | 0.02    | 7      
8     | 50   | 0.723    | 0.92     | 0.88    | 1.12     | 0.74    | 0.14    | 0.06    | 8      
9     | 50   | 0.750    | 1.00     | 0.96    | 1.02     | 0.76    | 0.14    | 0.05    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.526
Confusion matrix (rows=true, cols=pred):
 [[38  0  0  0  0  6  0  0  6  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 1  0 44  0  0  0  0  0  5  0]
 [ 0  0  1 27  0  1  0  0 21  0]
 [ 0  0  0  0  1  0  0  0 31 18]
 [ 0  0  1  0  0 16  0  0 33  0]
 [ 0  0  6  0  0  0 28  0 16  0]
 [ 0  0  0  0  0  0  0 18  7 25]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  9 41]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.76      0.85        50
           1       0.00      0.00      0.00        50
           2       0.85      0.88      0.86        50
           3       1.00      0.54      0.70        50
           4       1.00      0.02      0.04        50
           5       0.70      0.32      0.44        50
           6       1.00      0.56      0.72        50
           7       1.00      0.36      0.53        50
           8       0.22      1.00      0.36        50
           9       0.49      0.82      0.61        50

    accuracy                           0.53       500
   macro avg       0.72      0.53      0.51       500
weighted avg       0.72      0.53      0.51       500

Generation time: 1.0254 seconds
Coverage=0.226
Avg set size=0.23
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.431    | 0.44     | 0.44    | 1.42     | 0.45    | 0.21    | 0.14    | 0      
1     | 50   | 0.132    | 0.02     | 0.00    | 2.74     | 0.34    | 0.14    | 0.12    | 8      
2     | 50   | 0.466    | 0.62     | 0.62    | 1.16     | 0.48    | 0.21    | 0.10    | 2      
3     | 50   | 0.297    | 0.12     | 0.10    | 1.60     | 0.34    | 0.25    | 0.17    | 3      
4     | 50   | 0.161    | 0.00     | 0.00    | 3.02     | 0.27    | 0.21    | 0.16    | 8      
5     | 50   | 0.292    | 0.02     | 0.02    | 1.72     | 0.35    | 0.27    | 0.11    | 8      
6     | 50   | 0.284    | 0.10     | 0.10    | 1.68     | 0.32    | 0.22    | 0.16    | 6      
7     | 50   | 0.279    | 0.06     | 0.06    | 1.90     | 0.32    | 0.26    | 0.15    | 9      
8     | 50   | 0.485    | 0.82     | 0.82    | 1.00     | 0.48    | 0.16    | 0.11    | 8      
9     | 50   | 0.331    | 0.10     | 0.10    | 1.18     | 0.35    | 0.22    | 0.13    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0402 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.070    | 0.00     | 0.00    | 5.40     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.039    | 0.00     | 0.00    | 8.66     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.162    | 0.00     | 0.00    | 2.28     | 0.34    | 0.17    | 0.14    | 8      
3     | 50   | 0.116    | 0.00     | 0.00    | 3.58     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.10     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.183    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.11    | 8      
6     | 50   | 0.071    | 0.00     | 0.00    | 5.68     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 5.92     | 0.32    | 0.15    | 0.14    | 8      
8     | 50   | 0.374    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.141    | 0.00     | 0.00    | 2.68     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=12 | std=2.0 ---

== PCA [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.592
Confusion matrix (rows=true, cols=pred):
 [[25  0  8  1  0 10  1  1  3  1]
 [ 0 35  1  4  0  1  0  1  8  0]
 [ 2  1 27  3  0 11  2  1  2  1]
 [ 3  2 11 24  0  6  0  0  4  0]
 [ 0  0  1  1 29  5  0  4  5  5]
 [ 1  0  6  6  4 28  0  0  5  0]
 [ 0  1  3  0  4  5 35  0  1  1]
 [ 0  1  4  0  1  0  0 37  0  7]
 [ 0  1  9  3  1  7  3  0 23  3]
 [ 0  0  2  0  0  3  1  5  6 33]]
Classification report:
               precision    recall  f1-score   support

           0       0.81      0.50      0.62        50
           1       0.85      0.70      0.77        50
           2       0.38      0.54      0.44        50
           3       0.57      0.48      0.52        50
           4       0.74      0.58      0.65        50
           5       0.37      0.56      0.44        50
           6       0.83      0.70      0.76        50
           7       0.76      0.74      0.75        50
           8       0.40      0.46      0.43        50
           9       0.65      0.66      0.65        50

    accuracy                           0.59       500
   macro avg       0.64      0.59      0.60       500
weighted avg       0.64      0.59      0.60       500

Generation time: 0.6941 seconds
Coverage=0.536
Avg set size=0.81
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.366    | 0.80     | 0.42    | 2.20     | 0.56    | 0.23    | 0.10    | 0      
1     | 50   | 0.514    | 0.82     | 0.60    | 1.46     | 0.58    | 0.19    | 0.10    | 1      
2     | 50   | 0.459    | 0.80     | 0.50    | 2.60     | 0.65    | 0.16    | 0.08    | 2      
3     | 50   | 0.437    | 0.80     | 0.48    | 2.24     | 0.59    | 0.20    | 0.09    | 3      
4     | 50   | 0.471    | 0.74     | 0.52    | 1.86     | 0.62    | 0.17    | 0.09    | 4      
5     | 50   | 0.473    | 0.84     | 0.50    | 1.84     | 0.60    | 0.20    | 0.09    | 5      
6     | 50   | 0.537    | 0.80     | 0.66    | 1.54     | 0.61    | 0.19    | 0.08    | 6      
7     | 50   | 0.583    | 0.90     | 0.70    | 1.32     | 0.65    | 0.19    | 0.07    | 7      
8     | 50   | 0.382    | 0.78     | 0.38    | 1.84     | 0.56    | 0.23    | 0.10    | 8      
9     | 50   | 0.489    | 0.84     | 0.60    | 1.64     | 0.60    | 0.20    | 0.08    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.702
Confusion matrix (rows=true, cols=pred):
 [[39  0  2  1  0  6  0  0  2  0]
 [ 0 42  0  0  0  0  0  0  8  0]
 [ 0  0 46  0  0  0  0  0  4  0]
 [ 0  0  6 24  0  8  0  0 12  0]
 [ 1  0  1  0 21  1  0  0 14 12]
 [ 0  0  4  0  1 31  0  0 14  0]
 [ 1  0  7  0  0  2 31  0  9  0]
 [ 0  0  6  0  0  6  0 27  6  5]
 [ 0  0  1  0  0  1  0  0 47  1]
 [ 0  0  1  0  0  1  0  0  5 43]]
Classification report:
               precision    recall  f1-score   support

           0       0.95      0.78      0.86        50
           1       1.00      0.84      0.91        50
           2       0.62      0.92      0.74        50
           3       0.96      0.48      0.64        50
           4       0.95      0.42      0.58        50
           5       0.55      0.62      0.58        50
           6       1.00      0.62      0.77        50
           7       1.00      0.54      0.70        50
           8       0.39      0.94      0.55        50
           9       0.70      0.86      0.77        50

    accuracy                           0.70       500
   macro avg       0.81      0.70      0.71       500
weighted avg       0.81      0.70      0.71       500

Generation time: 1.7249 seconds
Coverage=0.464
Avg set size=0.50
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.489    | 0.56     | 0.56    | 1.46     | 0.51    | 0.20    | 0.12    | 0      
1     | 50   | 0.509    | 0.66     | 0.62    | 1.20     | 0.53    | 0.20    | 0.10    | 1      
2     | 50   | 0.485    | 0.66     | 0.66    | 1.12     | 0.49    | 0.18    | 0.11    | 2      
3     | 50   | 0.321    | 0.34     | 0.22    | 2.04     | 0.41    | 0.23    | 0.15    | 3      
4     | 50   | 0.265    | 0.24     | 0.18    | 2.40     | 0.34    | 0.24    | 0.14    | 4      
5     | 50   | 0.347    | 0.34     | 0.30    | 1.44     | 0.38    | 0.25    | 0.13    | 5      
6     | 50   | 0.379    | 0.38     | 0.34    | 1.80     | 0.44    | 0.19    | 0.12    | 6      
7     | 50   | 0.285    | 0.38     | 0.30    | 2.96     | 0.37    | 0.26    | 0.12    | 7      
8     | 50   | 0.590    | 0.86     | 0.86    | 1.06     | 0.59    | 0.15    | 0.09    | 8      
9     | 50   | 0.468    | 0.60     | 0.60    | 1.40     | 0.49    | 0.17    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.718
Confusion matrix (rows=true, cols=pred):
 [[39  0  2  1  0  7  0  0  1  0]
 [ 0 45  0  0  0  0  0  0  5  0]
 [ 0  0 40  0  0  1  0  0  9  0]
 [ 0  0  7 26  0  6  0  0 11  0]
 [ 2  0  2  1 26  1  1  0  8  9]
 [ 0  0  5  0  0 34  0  0 11  0]
 [ 2  0  3  0  0  2 36  0  7  0]
 [ 0  0  5  0  0  6  0 23  7  9]
 [ 0  0  0  0  0  2  0  0 47  1]
 [ 0  0  0  0  0  2  1  0  4 43]]
Classification report:
               precision    recall  f1-score   support

           0       0.91      0.78      0.84        50
           1       1.00      0.90      0.95        50
           2       0.62      0.80      0.70        50
           3       0.93      0.52      0.67        50
           4       1.00      0.52      0.68        50
           5       0.56      0.68      0.61        50
           6       0.95      0.72      0.82        50
           7       1.00      0.46      0.63        50
           8       0.43      0.94      0.59        50
           9       0.69      0.86      0.77        50

    accuracy                           0.72       500
   macro avg       0.81      0.72      0.73       500
weighted avg       0.81      0.72      0.73       500

Generation time: 1.0353 seconds
Coverage=0.512
Avg set size=0.55
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.492    | 0.60     | 0.60    | 1.48     | 0.51    | 0.19    | 0.12    | 0      
1     | 50   | 0.632    | 0.86     | 0.80    | 1.16     | 0.66    | 0.15    | 0.08    | 1      
2     | 50   | 0.446    | 0.54     | 0.54    | 1.22     | 0.46    | 0.20    | 0.12    | 2      
3     | 50   | 0.332    | 0.42     | 0.30    | 2.02     | 0.42    | 0.22    | 0.15    | 3      
4     | 50   | 0.299    | 0.36     | 0.28    | 2.48     | 0.38    | 0.24    | 0.13    | 4      
5     | 50   | 0.350    | 0.30     | 0.26    | 1.40     | 0.39    | 0.25    | 0.12    | 5      
6     | 50   | 0.425    | 0.54     | 0.48    | 1.84     | 0.48    | 0.18    | 0.12    | 6      
7     | 50   | 0.292    | 0.34     | 0.30    | 3.04     | 0.37    | 0.26    | 0.11    | 7      
8     | 50   | 0.634    | 0.92     | 0.90    | 1.08     | 0.64    | 0.14    | 0.08    | 8      
9     | 50   | 0.485    | 0.66     | 0.66    | 1.48     | 0.51    | 0.18    | 0.11    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.628
Confusion matrix (rows=true, cols=pred):
 [[33  0  1  0  0  8  0  0  8  0]
 [ 0 45  0  0  0  0  0  0  5  0]
 [ 0  0 31  0  0  3  0  0 16  0]
 [ 0  0  5 21  0  8  0  0 16  0]
 [ 0  0  3  0 23  2  0  0 14  8]
 [ 0  0  3  0  1 31  0  0 15  0]
 [ 1  0  5  0  0  4 28  0 12  0]
 [ 0  0  5  0  0  5  0 17 15  8]
 [ 0  0  0  0  0  2  0  0 48  0]
 [ 0  0  1  0  0  2  0  0 10 37]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.66      0.79        50
           1       1.00      0.90      0.95        50
           2       0.57      0.62      0.60        50
           3       1.00      0.42      0.59        50
           4       0.96      0.46      0.62        50
           5       0.48      0.62      0.54        50
           6       1.00      0.56      0.72        50
           7       1.00      0.34      0.51        50
           8       0.30      0.96      0.46        50
           9       0.70      0.74      0.72        50

    accuracy                           0.63       500
   macro avg       0.80      0.63      0.65       500
weighted avg       0.80      0.63      0.65       500

Generation time: 1.0672 seconds
Coverage=0.374
Avg set size=0.41
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.377    | 0.32     | 0.32    | 1.82     | 0.42    | 0.21    | 0.14    | 0      
1     | 50   | 0.658    | 0.86     | 0.78    | 1.14     | 0.68    | 0.15    | 0.07    | 1      
2     | 50   | 0.304    | 0.18     | 0.16    | 1.56     | 0.35    | 0.24    | 0.13    | 2      
3     | 50   | 0.278    | 0.26     | 0.20    | 2.56     | 0.37    | 0.23    | 0.16    | 3      
4     | 50   | 0.276    | 0.34     | 0.22    | 2.90     | 0.38    | 0.23    | 0.13    | 4      
5     | 50   | 0.318    | 0.20     | 0.16    | 1.44     | 0.35    | 0.25    | 0.12    | 5      
6     | 50   | 0.320    | 0.36     | 0.32    | 2.30     | 0.40    | 0.18    | 0.14    | 6      
7     | 50   | 0.222    | 0.20     | 0.20    | 4.12     | 0.35    | 0.24    | 0.13    | 7      
8     | 50   | 0.607    | 0.86     | 0.86    | 1.04     | 0.61    | 0.14    | 0.08    | 8      
9     | 50   | 0.428    | 0.52     | 0.52    | 1.88     | 0.47    | 0.17    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9524 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.114    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.089    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.043    | 0.00     | 0.00    | 6.98     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.162    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.050    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.043    | 0.00     | 0.00    | 7.46     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.880
Confusion matrix (rows=true, cols=pred):
 [[38  0  6  0  2  2  0  0  2  0]
 [ 0 48  1  0  0  0  0  0  1  0]
 [ 1  1 45  0  0  1  0  0  2  0]
 [ 0  0  1 48  0  1  0  0  0  0]
 [ 0  0  1  0 41  0  0  1  3  4]
 [ 0  0  3  2  1 38  1  0  4  1]
 [ 0  0  0  0  0  2 48  0  0  0]
 [ 0  0  0  0  0  0  0 45  1  4]
 [ 0  2  1  0  0  1  1  0 45  0]
 [ 0  0  0  3  0  1  0  2  0 44]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.76      0.85        50
           1       0.94      0.96      0.95        50
           2       0.78      0.90      0.83        50
           3       0.91      0.96      0.93        50
           4       0.93      0.82      0.87        50
           5       0.83      0.76      0.79        50
           6       0.96      0.96      0.96        50
           7       0.94      0.90      0.92        50
           8       0.78      0.90      0.83        50
           9       0.83      0.88      0.85        50

    accuracy                           0.88       500
   macro avg       0.89      0.88      0.88       500
weighted avg       0.89      0.88      0.88       500

Generation time: 0.9273 seconds
Coverage=0.830
Avg set size=0.89
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.708    | 0.92     | 0.76    | 1.76     | 0.77    | 0.12    | 0.05    | 0      
1     | 50   | 0.587    | 0.90     | 0.88    | 1.06     | 0.59    | 0.16    | 0.09    | 1      
2     | 50   | 0.861    | 0.94     | 0.90    | 1.18     | 0.88    | 0.05    | 0.02    | 2      
3     | 50   | 0.779    | 0.94     | 0.92    | 1.08     | 0.78    | 0.10    | 0.05    | 3      
4     | 50   | 0.572    | 0.82     | 0.74    | 1.32     | 0.61    | 0.18    | 0.08    | 4      
5     | 50   | 0.600    | 0.80     | 0.70    | 1.26     | 0.64    | 0.18    | 0.08    | 5      
6     | 50   | 0.797    | 0.90     | 0.86    | 1.10     | 0.82    | 0.10    | 0.03    | 6      
7     | 50   | 0.796    | 0.92     | 0.86    | 1.12     | 0.81    | 0.11    | 0.04    | 7      
8     | 50   | 0.640    | 0.88     | 0.84    | 1.20     | 0.66    | 0.16    | 0.08    | 8      
9     | 50   | 0.639    | 0.84     | 0.84    | 1.20     | 0.66    | 0.17    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=4.1806e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.1478 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 7.94     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=1.095e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.848
Confusion matrix (rows=true, cols=pred):
 [[33  0  2  1  1  8  1  0  3  1]
 [ 0 49  0  1  0  0  0  0  0  0]
 [ 0  0 45  2  0  0  0  1  2  0]
 [ 0  1  0 48  0  0  0  1  0  0]
 [ 0  1  0  0 38  1  0  1  0  9]
 [ 0  0  1  3  2 37  1  0  5  1]
 [ 0  0  0  0  0  4 46  0  0  0]
 [ 0  1  1  0  0  0  0 46  0  2]
 [ 0  3  1  3  0  4  1  0 38  0]
 [ 0  0  0  2  0  0  0  4  0 44]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.66      0.80        50
           1       0.89      0.98      0.93        50
           2       0.90      0.90      0.90        50
           3       0.80      0.96      0.87        50
           4       0.93      0.76      0.84        50
           5       0.69      0.74      0.71        50
           6       0.94      0.92      0.93        50
           7       0.87      0.92      0.89        50
           8       0.79      0.76      0.78        50
           9       0.77      0.88      0.82        50

    accuracy                           0.85       500
   macro avg       0.86      0.85      0.85       500
weighted avg       0.86      0.85      0.85       500

Generation time: 1.1599 seconds
Coverage=0.834
Avg set size=0.96
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.629    | 0.98     | 0.68    | 1.64     | 0.79    | 0.12    | 0.05    | 0      
1     | 50   | 0.739    | 0.92     | 0.92    | 1.02     | 0.74    | 0.11    | 0.06    | 1      
2     | 50   | 0.863    | 0.96     | 0.90    | 1.16     | 0.89    | 0.05    | 0.03    | 2      
3     | 50   | 0.888    | 0.98     | 0.96    | 1.08     | 0.90    | 0.06    | 0.02    | 3      
4     | 50   | 0.657    | 0.92     | 0.76    | 1.42     | 0.72    | 0.15    | 0.06    | 4      
5     | 50   | 0.670    | 0.98     | 0.74    | 1.38     | 0.75    | 0.15    | 0.05    | 5      
6     | 50   | 0.864    | 0.96     | 0.90    | 1.08     | 0.89    | 0.07    | 0.02    | 6      
7     | 50   | 0.790    | 0.96     | 0.88    | 1.28     | 0.82    | 0.12    | 0.03    | 7      
8     | 50   | 0.647    | 0.98     | 0.76    | 1.34     | 0.72    | 0.16    | 0.06    | 8      
9     | 50   | 0.706    | 0.92     | 0.84    | 1.14     | 0.73    | 0.15    | 0.06    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.514
Confusion matrix (rows=true, cols=pred):
 [[34  0  1  0  0  8  0  0  7  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 1  0 40  0  0  0  0  0  9  0]
 [ 0  0  2 27  0  2  0  0 19  0]
 [ 0  0  2  0  2  0  0  0 28 18]
 [ 0  0  2  1  0 21  0  1 25  0]
 [ 0  0  8  0  0  4 23  0 15  0]
 [ 0  0  0  0  0  0  0 21  9 20]
 [ 0  0  0  1  0  0  0  0 49  0]
 [ 0  0  0  0  0  0  0  0 10 40]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.68      0.80        50
           1       0.00      0.00      0.00        50
           2       0.73      0.80      0.76        50
           3       0.93      0.54      0.68        50
           4       1.00      0.04      0.08        50
           5       0.60      0.42      0.49        50
           6       1.00      0.46      0.63        50
           7       0.95      0.42      0.58        50
           8       0.22      0.98      0.36        50
           9       0.51      0.80      0.62        50

    accuracy                           0.51       500
   macro avg       0.69      0.51      0.50       500
weighted avg       0.69      0.51      0.50       500

Generation time: 1.0559 seconds
Coverage=0.228
Avg set size=0.25
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.437    | 0.46     | 0.44    | 1.64     | 0.48    | 0.20    | 0.13    | 0      
1     | 50   | 0.132    | 0.04     | 0.00    | 2.98     | 0.34    | 0.15    | 0.12    | 8      
2     | 50   | 0.457    | 0.56     | 0.54    | 1.32     | 0.48    | 0.19    | 0.11    | 2      
3     | 50   | 0.298    | 0.14     | 0.12    | 1.64     | 0.35    | 0.24    | 0.17    | 3      
4     | 50   | 0.158    | 0.00     | 0.00    | 3.24     | 0.27    | 0.21    | 0.16    | 8      
5     | 50   | 0.283    | 0.14     | 0.08    | 1.76     | 0.35    | 0.26    | 0.12    | 8      
6     | 50   | 0.281    | 0.12     | 0.12    | 2.00     | 0.35    | 0.22    | 0.15    | 6      
7     | 50   | 0.282    | 0.20     | 0.14    | 2.04     | 0.34    | 0.25    | 0.14    | 7      
8     | 50   | 0.471    | 0.72     | 0.72    | 1.02     | 0.47    | 0.17    | 0.11    | 8      
9     | 50   | 0.329    | 0.12     | 0.12    | 1.32     | 0.35    | 0.22    | 0.13    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0361 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.071    | 0.00     | 0.00    | 5.44     | 0.32    | 0.19    | 0.12    | 8      
1     | 50   | 0.039    | 0.00     | 0.00    | 8.58     | 0.35    | 0.14    | 0.12    | 8      
2     | 50   | 0.163    | 0.00     | 0.00    | 2.30     | 0.33    | 0.17    | 0.14    | 8      
3     | 50   | 0.116    | 0.00     | 0.00    | 3.64     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.20     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.183    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.11    | 8      
6     | 50   | 0.071    | 0.00     | 0.00    | 5.52     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 5.88     | 0.32    | 0.15    | 0.14    | 8      
8     | 50   | 0.374    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.141    | 0.00     | 0.00    | 2.66     | 0.33    | 0.15    | 0.14    | 8      


=== Diffusion steps: T=1000 ===

--- latent_size=8 | std=1.0 ---

== PCA [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.762
Confusion matrix (rows=true, cols=pred):
 [[31  0  1  2  0 14  0  0  2  0]
 [ 0 31  0  0  0  1  0  0 18  0]
 [ 0  0 37  3  2  2  1  0  4  1]
 [ 0  0  7 29  0  6  0  1  7  0]
 [ 0  0  0  0 44  1  0  0  2  3]
 [ 1  0  2  0  0 42  0  0  5  0]
 [ 0  1  0  0  0  5 41  0  3  0]
 [ 0  0  0  0  1  0  0 42  0  7]
 [ 0  0  0  2  0  6  0  0 41  1]
 [ 0  0  0  0  1  0  0  0  6 43]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.62      0.76        50
           1       0.97      0.62      0.76        50
           2       0.79      0.74      0.76        50
           3       0.81      0.58      0.67        50
           4       0.92      0.88      0.90        50
           5       0.55      0.84      0.66        50
           6       0.98      0.82      0.89        50
           7       0.98      0.84      0.90        50
           8       0.47      0.82      0.59        50
           9       0.78      0.86      0.82        50

    accuracy                           0.76       500
   macro avg       0.82      0.76      0.77       500
weighted avg       0.82      0.76      0.77       500

Generation time: 0.7487 seconds
Coverage=0.582
Avg set size=0.69
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.370    | 0.56     | 0.40    | 1.74     | 0.46    | 0.22    | 0.12    | 0      
1     | 50   | 0.362    | 0.56     | 0.36    | 1.46     | 0.45    | 0.24    | 0.12    | 1      
2     | 50   | 0.479    | 0.72     | 0.62    | 1.50     | 0.54    | 0.16    | 0.10    | 2      
3     | 50   | 0.376    | 0.54     | 0.36    | 1.86     | 0.47    | 0.22    | 0.12    | 3      
4     | 50   | 0.532    | 0.76     | 0.70    | 1.22     | 0.56    | 0.18    | 0.11    | 4      
5     | 50   | 0.525    | 0.68     | 0.60    | 1.22     | 0.55    | 0.19    | 0.09    | 5      
6     | 50   | 0.497    | 0.72     | 0.68    | 1.26     | 0.52    | 0.18    | 0.10    | 6      
7     | 50   | 0.546    | 0.84     | 0.76    | 1.20     | 0.58    | 0.22    | 0.09    | 7      
8     | 50   | 0.498    | 0.70     | 0.58    | 1.18     | 0.53    | 0.20    | 0.09    | 8      
9     | 50   | 0.547    | 0.78     | 0.76    | 1.28     | 0.57    | 0.18    | 0.10    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.898
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  3 40  0  0  0  0  7  0]
 [ 0  0  0  0 33  0  0  0  4 13]
 [ 0  0  0  0  0 44  0  0  6  0]
 [ 0  0  2  0  0  0 48  0  0  0]
 [ 0  0  0  0  0  0  0 35  5 10]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  1 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       0.91      1.00      0.95        50
           3       1.00      0.80      0.89        50
           4       1.00      0.66      0.80        50
           5       1.00      0.88      0.94        50
           6       1.00      0.96      0.98        50
           7       1.00      0.70      0.82        50
           8       0.68      1.00      0.81        50
           9       0.68      0.98      0.80        50

    accuracy                           0.90       500
   macro avg       0.93      0.90      0.90       500
weighted avg       0.93      0.90      0.90       500

Generation time: 1.4117 seconds
Coverage=0.538
Avg set size=0.54
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.539    | 0.94     | 0.94    | 1.00     | 0.54    | 0.20    | 0.11    | 0      
1     | 50   | 0.579    | 1.00     | 1.00    | 1.00     | 0.58    | 0.17    | 0.08    | 1      
2     | 50   | 0.491    | 0.82     | 0.82    | 1.00     | 0.49    | 0.20    | 0.09    | 2      
3     | 50   | 0.348    | 0.18     | 0.18    | 1.28     | 0.35    | 0.25    | 0.18    | 3      
4     | 50   | 0.302    | 0.02     | 0.02    | 1.38     | 0.31    | 0.26    | 0.18    | 4      
5     | 50   | 0.374    | 0.10     | 0.10    | 1.12     | 0.38    | 0.32    | 0.08    | 5      
6     | 50   | 0.381    | 0.32     | 0.32    | 1.04     | 0.38    | 0.18    | 0.14    | 6      
7     | 50   | 0.311    | 0.18     | 0.18    | 1.36     | 0.32    | 0.27    | 0.14    | 7      
8     | 50   | 0.602    | 1.00     | 1.00    | 1.00     | 0.60    | 0.12    | 0.08    | 8      
9     | 50   | 0.501    | 0.82     | 0.82    | 1.02     | 0.50    | 0.16    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.928
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  1 39  0  1  0  0  9  0]
 [ 0  0  0  0 42  0  0  0  5  3]
 [ 0  0  0  0  0 49  0  0  1  0]
 [ 0  0  3  0  0  0 47  0  0  0]
 [ 0  0  0  0  0  0  0 38  4  8]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  1 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       0.93      1.00      0.96        50
           3       1.00      0.78      0.88        50
           4       1.00      0.84      0.91        50
           5       0.98      0.98      0.98        50
           6       1.00      0.94      0.97        50
           7       1.00      0.76      0.86        50
           8       0.71      1.00      0.83        50
           9       0.82      0.98      0.89        50

    accuracy                           0.93       500
   macro avg       0.94      0.93      0.93       500
weighted avg       0.94      0.93      0.93       500

Generation time: 0.9326 seconds
Coverage=0.520
Avg set size=0.52
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.518    | 0.84     | 0.84    | 1.00     | 0.52    | 0.21    | 0.11    | 0      
1     | 50   | 0.739    | 1.00     | 1.00    | 1.00     | 0.74    | 0.11    | 0.06    | 1      
2     | 50   | 0.446    | 0.60     | 0.60    | 1.00     | 0.45    | 0.22    | 0.10    | 2      
3     | 50   | 0.336    | 0.08     | 0.08    | 1.36     | 0.35    | 0.25    | 0.19    | 3      
4     | 50   | 0.348    | 0.16     | 0.16    | 1.18     | 0.35    | 0.26    | 0.15    | 4      
5     | 50   | 0.387    | 0.18     | 0.18    | 1.02     | 0.39    | 0.31    | 0.08    | 5      
6     | 50   | 0.375    | 0.34     | 0.34    | 1.06     | 0.38    | 0.18    | 0.15    | 6      
7     | 50   | 0.316    | 0.18     | 0.18    | 1.46     | 0.33    | 0.27    | 0.13    | 7      
8     | 50   | 0.629    | 1.00     | 1.00    | 1.00     | 0.63    | 0.12    | 0.08    | 8      
9     | 50   | 0.528    | 0.82     | 0.82    | 1.02     | 0.53    | 0.15    | 0.11    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.726
Confusion matrix (rows=true, cols=pred):
 [[41  0  0  0  0  9  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 38  0  0  0  0  0 12  0]
 [ 0  0  0 18  0  2  0  0 30  0]
 [ 0  0  0  0 32  0  0  0 12  6]
 [ 0  0  0  0  0 37  0  0 13  0]
 [ 0  0  4  0  0  4 32  0 10  0]
 [ 0  0  0  0  0  0  0 15 15 20]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.82      0.90        50
           1       1.00      1.00      1.00        50
           2       0.90      0.76      0.83        50
           3       1.00      0.36      0.53        50
           4       1.00      0.64      0.78        50
           5       0.71      0.74      0.73        50
           6       1.00      0.64      0.78        50
           7       1.00      0.30      0.46        50
           8       0.35      1.00      0.52        50
           9       0.66      1.00      0.79        50

    accuracy                           0.73       500
   macro avg       0.86      0.73      0.73       500
weighted avg       0.86      0.73      0.73       500

Generation time: 0.9517 seconds
Coverage=0.298
Avg set size=0.30
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.329    | 0.12     | 0.12    | 1.26     | 0.34    | 0.24    | 0.17    | 0      
1     | 50   | 0.816    | 1.00     | 1.00    | 1.00     | 0.82    | 0.07    | 0.04    | 1      
2     | 50   | 0.297    | 0.00     | 0.00    | 1.24     | 0.30    | 0.27    | 0.15    | 2      
3     | 50   | 0.264    | 0.00     | 0.00    | 1.90     | 0.31    | 0.26    | 0.20    | 8      
4     | 50   | 0.293    | 0.08     | 0.08    | 1.52     | 0.31    | 0.24    | 0.17    | 4      
5     | 50   | 0.347    | 0.02     | 0.02    | 1.26     | 0.35    | 0.31    | 0.09    | 5      
6     | 50   | 0.269    | 0.06     | 0.06    | 1.70     | 0.28    | 0.20    | 0.18    | 6      
7     | 50   | 0.229    | 0.00     | 0.00    | 2.46     | 0.27    | 0.24    | 0.16    | 9      
8     | 50   | 0.587    | 0.98     | 0.98    | 1.00     | 0.59    | 0.13    | 0.08    | 8      
9     | 50   | 0.479    | 0.72     | 0.72    | 1.00     | 0.48    | 0.17    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.8546 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 7.98     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.990
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 50  0  0  0  0  0  0]
 [ 0  0  0  0 48  0  0  0  0  2]
 [ 0  0  0  0  0 49  0  0  1  0]
 [ 0  0  0  0  0  0 50  0  0  0]
 [ 0  0  0  0  0  0  0 48  0  2]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       1.00      1.00      1.00        50
           3       1.00      1.00      1.00        50
           4       1.00      0.96      0.98        50
           5       1.00      0.98      0.99        50
           6       1.00      1.00      1.00        50
           7       1.00      0.96      0.98        50
           8       0.98      1.00      0.99        50
           9       0.93      1.00      0.96        50

    accuracy                           0.99       500
   macro avg       0.99      0.99      0.99       500
weighted avg       0.99      0.99      0.99       500

Generation time: 1.3938 seconds
Coverage=0.950
Avg set size=0.96
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.838    | 0.92     | 0.92    | 1.00     | 0.84    | 0.09    | 0.03    | 0      
1     | 50   | 0.503    | 0.88     | 0.88    | 1.00     | 0.50    | 0.18    | 0.10    | 1      
2     | 50   | 0.936    | 1.00     | 1.00    | 1.00     | 0.94    | 0.03    | 0.02    | 2      
3     | 50   | 0.753    | 0.98     | 0.98    | 1.00     | 0.75    | 0.11    | 0.07    | 3      
4     | 50   | 0.547    | 0.86     | 0.82    | 1.04     | 0.55    | 0.24    | 0.08    | 4      
5     | 50   | 0.566    | 0.96     | 0.94    | 1.02     | 0.57    | 0.28    | 0.06    | 5      
6     | 50   | 0.875    | 1.00     | 1.00    | 1.00     | 0.87    | 0.05    | 0.02    | 6      
7     | 50   | 0.791    | 1.04     | 0.98    | 1.04     | 0.79    | 0.16    | 0.02    | 7      
8     | 50   | 0.734    | 0.98     | 0.98    | 1.00     | 0.73    | 0.10    | 0.06    | 8      
9     | 50   | 0.663    | 1.00     | 1.00    | 1.00     | 0.66    | 0.16    | 0.09    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=3.58354e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.5642 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=4.30212e-09): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.628
Confusion matrix (rows=true, cols=pred):
 [[43  0  0  0  0  5  0  0  2  0]
 [ 0 39  0  0  0  0  0  0 11  0]
 [ 0  0 38  3  0  3  3  0  3  0]
 [ 0  0  3 33  0  7  0  0  7  0]
 [ 0  0  0  0  0  0  0  0  1 49]
 [ 0  0  5  1  0 11  0  0 32  1]
 [ 0  0 10  0  0  4 25  0 11  0]
 [ 0  0  0  0  0  0  0 33  0 17]
 [ 0  0  1  4  0  0  0  0 45  0]
 [ 0  0  0  0  0  0  0  2  1 47]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.86      0.92        50
           1       1.00      0.78      0.88        50
           2       0.67      0.76      0.71        50
           3       0.80      0.66      0.73        50
           4       0.00      0.00      0.00        50
           5       0.37      0.22      0.27        50
           6       0.89      0.50      0.64        50
           7       0.94      0.66      0.78        50
           8       0.40      0.90      0.55        50
           9       0.41      0.94      0.57        50

    accuracy                           0.63       500
   macro avg       0.65      0.63      0.61       500
weighted avg       0.65      0.63      0.61       500

Generation time: 1.0679 seconds
Coverage=0.472
Avg set size=0.59
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.694    | 0.86     | 0.80    | 1.26     | 0.72    | 0.13    | 0.07    | 0      
1     | 50   | 0.373    | 0.26     | 0.26    | 1.22     | 0.38    | 0.27    | 0.10    | 1      
2     | 50   | 0.601    | 0.84     | 0.70    | 1.42     | 0.67    | 0.15    | 0.08    | 2      
3     | 50   | 0.501    | 0.64     | 0.56    | 1.98     | 0.59    | 0.18    | 0.10    | 3      
4     | 50   | 0.167    | 0.48     | 0.00    | 2.76     | 0.42    | 0.19    | 0.14    | 9      
5     | 50   | 0.251    | 0.16     | 0.12    | 2.30     | 0.37    | 0.25    | 0.12    | 8      
6     | 50   | 0.448    | 0.58     | 0.48    | 2.40     | 0.55    | 0.20    | 0.10    | 6      
7     | 50   | 0.536    | 0.94     | 0.64    | 1.44     | 0.63    | 0.24    | 0.06    | 7      
8     | 50   | 0.399    | 0.38     | 0.38    | 1.12     | 0.40    | 0.22    | 0.12    | 8      
9     | 50   | 0.562    | 0.80     | 0.78    | 1.08     | 0.56    | 0.21    | 0.10    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.590
Confusion matrix (rows=true, cols=pred):
 [[49  0  0  0  0  1  0  0  0  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 29  0  0  0  0 21  0]
 [ 0  0  0  0  0  0  0  0 45  5]
 [ 0  0  0  0  0  5  0  0 45  0]
 [ 0  0  1  0  0  0 44  0  5  0]
 [ 0  0  0  0  0  0  0 24  0 26]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  6 44]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.98      0.99        50
           1       0.00      0.00      0.00        50
           2       0.98      1.00      0.99        50
           3       1.00      0.58      0.73        50
           4       0.00      0.00      0.00        50
           5       0.83      0.10      0.18        50
           6       1.00      0.88      0.94        50
           7       1.00      0.48      0.65        50
           8       0.23      1.00      0.37        50
           9       0.59      0.88      0.70        50

    accuracy                           0.59       500
   macro avg       0.66      0.59      0.55       500
weighted avg       0.66      0.59      0.55       500

Generation time: 0.9443 seconds
Coverage=0.222
Avg set size=0.22
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.423    | 0.44     | 0.44    | 1.02     | 0.42    | 0.23    | 0.16    | 0      
1     | 50   | 0.131    | 0.00     | 0.00    | 2.42     | 0.33    | 0.13    | 0.12    | 8      
2     | 50   | 0.498    | 0.78     | 0.78    | 1.00     | 0.50    | 0.21    | 0.08    | 2      
3     | 50   | 0.294    | 0.00     | 0.00    | 1.42     | 0.31    | 0.27    | 0.17    | 3      
4     | 50   | 0.169    | 0.00     | 0.00    | 3.00     | 0.26    | 0.22    | 0.17    | 8      
5     | 50   | 0.301    | 0.00     | 0.00    | 1.90     | 0.36    | 0.30    | 0.08    | 8      
6     | 50   | 0.299    | 0.00     | 0.00    | 1.14     | 0.31    | 0.21    | 0.17    | 6      
7     | 50   | 0.287    | 0.00     | 0.00    | 1.56     | 0.31    | 0.27    | 0.16    | 9      
8     | 50   | 0.505    | 1.00     | 1.00    | 1.00     | 0.51    | 0.14    | 0.10    | 8      
9     | 50   | 0.332    | 0.00     | 0.00    | 1.12     | 0.33    | 0.24    | 0.12    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9466 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.069    | 0.00     | 0.00    | 5.60     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.038    | 0.00     | 0.00    | 9.00     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.162    | 0.00     | 0.00    | 2.10     | 0.34    | 0.16    | 0.15    | 8      
3     | 50   | 0.117    | 0.00     | 0.00    | 3.50     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.02     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.184    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.10    | 8      
6     | 50   | 0.072    | 0.00     | 0.00    | 5.64     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 6.00     | 0.32    | 0.15    | 0.14    | 8      
8     | 50   | 0.373    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.141    | 0.00     | 0.00    | 2.84     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=8 | std=1.5 ---

== PCA [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.642
Confusion matrix (rows=true, cols=pred):
 [[24  0  3  3  0 14  0  2  3  1]
 [ 0 27  1  2  0  1  1  0 18  0]
 [ 0  0 36  5  2  2  1  0  4  0]
 [ 0  0  9 26  1  7  0  2  5  0]
 [ 0  0  1  0 38  3  2  0  3  3]
 [ 1  0  4  2  1 33  0  0  7  2]
 [ 0  1  2  1  3  5 34  0  4  0]
 [ 0  0  1  0  4  0  0 40  0  5]
 [ 0  1  1  6  0 12  0  0 28  2]
 [ 0  0  1  0  3  3  0  6  2 35]]
Classification report:
               precision    recall  f1-score   support

           0       0.96      0.48      0.64        50
           1       0.93      0.54      0.68        50
           2       0.61      0.72      0.66        50
           3       0.58      0.52      0.55        50
           4       0.73      0.76      0.75        50
           5       0.41      0.66      0.51        50
           6       0.89      0.68      0.77        50
           7       0.80      0.80      0.80        50
           8       0.38      0.56      0.45        50
           9       0.73      0.70      0.71        50

    accuracy                           0.64       500
   macro avg       0.70      0.64      0.65       500
weighted avg       0.70      0.64      0.65       500

Generation time: 0.6056 seconds
Coverage=0.528
Avg set size=0.72
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.346    | 0.66     | 0.38    | 2.10     | 0.51    | 0.22    | 0.11    | 0      
1     | 50   | 0.344    | 0.66     | 0.38    | 1.92     | 0.49    | 0.22    | 0.12    | 1      
2     | 50   | 0.515    | 0.78     | 0.62    | 1.78     | 0.62    | 0.17    | 0.08    | 2      
3     | 50   | 0.370    | 0.68     | 0.38    | 2.14     | 0.55    | 0.20    | 0.11    | 3      
4     | 50   | 0.484    | 0.72     | 0.58    | 1.52     | 0.56    | 0.19    | 0.11    | 4      
5     | 50   | 0.512    | 0.70     | 0.56    | 1.60     | 0.59    | 0.18    | 0.09    | 5      
6     | 50   | 0.452    | 0.68     | 0.54    | 1.66     | 0.54    | 0.20    | 0.11    | 6      
7     | 50   | 0.550    | 0.78     | 0.68    | 1.32     | 0.59    | 0.20    | 0.09    | 7      
8     | 50   | 0.428    | 0.74     | 0.48    | 1.66     | 0.56    | 0.19    | 0.10    | 8      
9     | 50   | 0.504    | 0.82     | 0.68    | 1.64     | 0.57    | 0.20    | 0.09    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.832
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  4 35  0  2  0  0  9  0]
 [ 0  0  0  0 28  0  0  0  8 14]
 [ 0  0  0  0  0 34  0  0 16  0]
 [ 0  0  5  0  0  0 43  0  2  0]
 [ 0  0  0  0  0  1  0 30  8 11]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  2 48]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      0.98      0.99        50
           2       0.84      0.98      0.91        50
           3       1.00      0.70      0.82        50
           4       1.00      0.56      0.72        50
           5       0.92      0.68      0.78        50
           6       1.00      0.86      0.92        50
           7       1.00      0.60      0.75        50
           8       0.52      1.00      0.68        50
           9       0.66      0.96      0.78        50

    accuracy                           0.83       500
   macro avg       0.89      0.83      0.84       500
weighted avg       0.89      0.83      0.84       500

Generation time: 0.9548 seconds
Coverage=0.522
Avg set size=0.53
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.534    | 0.78     | 0.78    | 1.00     | 0.53    | 0.20    | 0.12    | 0      
1     | 50   | 0.563    | 0.90     | 0.90    | 1.02     | 0.56    | 0.18    | 0.09    | 1      
2     | 50   | 0.483    | 0.76     | 0.76    | 1.02     | 0.48    | 0.20    | 0.10    | 2      
3     | 50   | 0.356    | 0.32     | 0.32    | 1.46     | 0.38    | 0.25    | 0.17    | 3      
4     | 50   | 0.299    | 0.10     | 0.10    | 1.60     | 0.32    | 0.25    | 0.16    | 4      
5     | 50   | 0.366    | 0.16     | 0.14    | 1.32     | 0.38    | 0.31    | 0.09    | 5      
6     | 50   | 0.374    | 0.32     | 0.32    | 1.22     | 0.39    | 0.18    | 0.14    | 6      
7     | 50   | 0.310    | 0.22     | 0.18    | 1.72     | 0.33    | 0.27    | 0.12    | 7      
8     | 50   | 0.584    | 0.96     | 0.96    | 1.00     | 0.58    | 0.13    | 0.09    | 8      
9     | 50   | 0.499    | 0.76     | 0.76    | 1.08     | 0.50    | 0.17    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.868
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  3 37  0  0  0  0 10  0]
 [ 0  0  0  0 37  0  0  0  9  4]
 [ 0  0  0  0  0 40  0  0 10  0]
 [ 0  0  5  0  0  1 42  0  2  0]
 [ 0  0  1  0  0  0  0 31  5 13]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  1  0  0  0  0  0  2 47]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       0.83      1.00      0.91        50
           3       1.00      0.74      0.85        50
           4       1.00      0.74      0.85        50
           5       0.98      0.80      0.88        50
           6       1.00      0.84      0.91        50
           7       1.00      0.62      0.77        50
           8       0.57      1.00      0.72        50
           9       0.73      0.94      0.82        50

    accuracy                           0.87       500
   macro avg       0.91      0.87      0.87       500
weighted avg       0.91      0.87      0.87       500

Generation time: 1.4383 seconds
Coverage=0.550
Avg set size=0.55
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.515    | 0.82     | 0.82    | 1.00     | 0.52    | 0.21    | 0.12    | 0      
1     | 50   | 0.720    | 1.00     | 1.00    | 1.00     | 0.72    | 0.12    | 0.06    | 1      
2     | 50   | 0.444    | 0.54     | 0.54    | 1.00     | 0.44    | 0.22    | 0.11    | 2      
3     | 50   | 0.341    | 0.26     | 0.26    | 1.46     | 0.37    | 0.25    | 0.18    | 3      
4     | 50   | 0.347    | 0.28     | 0.26    | 1.36     | 0.37    | 0.26    | 0.14    | 4      
5     | 50   | 0.379    | 0.24     | 0.24    | 1.20     | 0.38    | 0.30    | 0.09    | 5      
6     | 50   | 0.364    | 0.36     | 0.36    | 1.30     | 0.38    | 0.18    | 0.14    | 6      
7     | 50   | 0.320    | 0.30     | 0.28    | 1.80     | 0.34    | 0.27    | 0.12    | 7      
8     | 50   | 0.613    | 0.96     | 0.96    | 1.00     | 0.61    | 0.12    | 0.09    | 8      
9     | 50   | 0.524    | 0.78     | 0.78    | 1.10     | 0.53    | 0.17    | 0.11    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.700
Confusion matrix (rows=true, cols=pred):
 [[37  0  0  0  0 10  0  0  3  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 33  0  0  1  0  0 16  0]
 [ 0  0  0 25  0  5  0  0 20  0]
 [ 0  0  0  0 28  0  0  0 15  7]
 [ 0  0  0  0  0 34  0  0 16  0]
 [ 0  0  5  0  0  6 27  0 12  0]
 [ 0  0  0  0  0  1  0 18 18 13]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  2 48]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.74      0.85        50
           1       1.00      1.00      1.00        50
           2       0.87      0.66      0.75        50
           3       1.00      0.50      0.67        50
           4       1.00      0.56      0.72        50
           5       0.60      0.68      0.64        50
           6       1.00      0.54      0.70        50
           7       1.00      0.36      0.53        50
           8       0.33      1.00      0.50        50
           9       0.71      0.96      0.81        50

    accuracy                           0.70       500
   macro avg       0.85      0.70      0.72       500
weighted avg       0.85      0.70      0.72       500

Generation time: 0.9095 seconds
Coverage=0.324
Avg set size=0.33
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.329    | 0.20     | 0.20    | 1.48     | 0.35    | 0.23    | 0.17    | 0      
1     | 50   | 0.793    | 0.98     | 0.98    | 1.00     | 0.79    | 0.08    | 0.05    | 1      
2     | 50   | 0.298    | 0.00     | 0.00    | 1.34     | 0.31    | 0.26    | 0.15    | 2      
3     | 50   | 0.274    | 0.08     | 0.08    | 1.88     | 0.32    | 0.25    | 0.19    | 3      
4     | 50   | 0.294    | 0.16     | 0.16    | 1.78     | 0.33    | 0.24    | 0.15    | 4      
5     | 50   | 0.345    | 0.10     | 0.10    | 1.32     | 0.36    | 0.30    | 0.09    | 5      
6     | 50   | 0.264    | 0.12     | 0.12    | 2.12     | 0.30    | 0.20    | 0.17    | 6      
7     | 50   | 0.239    | 0.12     | 0.10    | 2.64     | 0.29    | 0.24    | 0.14    | 7      
8     | 50   | 0.571    | 0.88     | 0.88    | 1.00     | 0.57    | 0.14    | 0.09    | 8      
9     | 50   | 0.471    | 0.62     | 0.62    | 1.06     | 0.47    | 0.17    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.3208 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.049    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 7.82     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.956
Confusion matrix (rows=true, cols=pred):
 [[46  0  0  0  0  2  0  0  1  1]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  1 49  0  0  0  0  0  0]
 [ 0  0  0  0 41  0  0  0  0  9]
 [ 0  0  0  1  0 47  0  0  2  0]
 [ 0  0  0  0  0  0 50  0  0  0]
 [ 0  0  0  0  0  0  0 46  0  4]
 [ 0  0  1  0  0  0  0  0 49  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.92      0.96        50
           1       1.00      1.00      1.00        50
           2       0.96      1.00      0.98        50
           3       0.98      0.98      0.98        50
           4       1.00      0.82      0.90        50
           5       0.96      0.94      0.95        50
           6       1.00      1.00      1.00        50
           7       1.00      0.92      0.96        50
           8       0.94      0.98      0.96        50
           9       0.78      1.00      0.88        50

    accuracy                           0.96       500
   macro avg       0.96      0.96      0.96       500
weighted avg       0.96      0.96      0.96       500

Generation time: 0.8035 seconds
Coverage=0.902
Avg set size=0.92
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.782    | 0.92     | 0.90    | 1.18     | 0.79    | 0.11    | 0.04    | 0      
1     | 50   | 0.493    | 0.76     | 0.76    | 1.00     | 0.49    | 0.18    | 0.11    | 1      
2     | 50   | 0.916    | 1.00     | 1.00    | 1.00     | 0.92    | 0.04    | 0.02    | 2      
3     | 50   | 0.733    | 0.96     | 0.96    | 1.02     | 0.74    | 0.12    | 0.07    | 3      
4     | 50   | 0.547    | 0.76     | 0.70    | 1.18     | 0.56    | 0.22    | 0.08    | 4      
5     | 50   | 0.560    | 0.82     | 0.80    | 1.06     | 0.56    | 0.26    | 0.07    | 5      
6     | 50   | 0.856    | 0.98     | 0.98    | 1.00     | 0.86    | 0.06    | 0.03    | 6      
7     | 50   | 0.781    | 1.06     | 0.96    | 1.08     | 0.79    | 0.16    | 0.02    | 7      
8     | 50   | 0.707    | 0.98     | 0.96    | 1.02     | 0.71    | 0.12    | 0.07    | 8      
9     | 50   | 0.655    | 1.00     | 1.00    | 1.00     | 0.65    | 0.17    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=3.58354e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.6168 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=4.30212e-09): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.554
Confusion matrix (rows=true, cols=pred):
 [[35  0  2  0  0  7  0  0  6  0]
 [ 0 37  0  0  0  0  0  0 13  0]
 [ 1  0 34  3  0  3  6  0  3  0]
 [ 0  0  3 28  0  7  4  0  8  0]
 [ 0  0  0  0  0  0  0  0  2 48]
 [ 0  0  9  2  0 13  0  1 23  2]
 [ 0  0  9  1  0 10 23  0  7  0]
 [ 0  1  0  1  0  0  0 30  3 15]
 [ 0  0  3 10  0  1  0  0 36  0]
 [ 0  0  0  3  0  0  0  5  1 41]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.70      0.81        50
           1       0.97      0.74      0.84        50
           2       0.57      0.68      0.62        50
           3       0.58      0.56      0.57        50
           4       0.00      0.00      0.00        50
           5       0.32      0.26      0.29        50
           6       0.70      0.46      0.55        50
           7       0.83      0.60      0.70        50
           8       0.35      0.72      0.47        50
           9       0.39      0.82      0.53        50

    accuracy                           0.55       500
   macro avg       0.57      0.55      0.54       500
weighted avg       0.57      0.55      0.54       500

Generation time: 1.6045 seconds
Coverage=0.452
Avg set size=0.63
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.624    | 0.78     | 0.66    | 1.86     | 0.70    | 0.14    | 0.07    | 0      
1     | 50   | 0.376    | 0.34     | 0.34    | 1.26     | 0.40    | 0.25    | 0.11    | 1      
2     | 50   | 0.557    | 0.86     | 0.64    | 1.80     | 0.70    | 0.14    | 0.07    | 2      
3     | 50   | 0.500    | 0.76     | 0.54    | 2.76     | 0.63    | 0.17    | 0.08    | 3      
4     | 50   | 0.165    | 0.44     | 0.00    | 3.00     | 0.41    | 0.21    | 0.14    | 9      
5     | 50   | 0.257    | 0.32     | 0.18    | 2.70     | 0.40    | 0.24    | 0.12    | 8      
6     | 50   | 0.429    | 0.66     | 0.46    | 3.10     | 0.60    | 0.18    | 0.10    | 6      
7     | 50   | 0.517    | 0.96     | 0.60    | 1.76     | 0.65    | 0.21    | 0.07    | 7      
8     | 50   | 0.391    | 0.40     | 0.38    | 1.34     | 0.42    | 0.23    | 0.12    | 8      
9     | 50   | 0.558    | 0.78     | 0.72    | 1.26     | 0.58    | 0.20    | 0.09    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.566
Confusion matrix (rows=true, cols=pred):
 [[42  0  1  0  0  5  0  0  2  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 48  0  0  0  0  0  2  0]
 [ 0  0  0 29  0  0  0  0 21  0]
 [ 0  0  0  0  2  0  0  0 41  7]
 [ 0  0  0  0  0  7  0  0 43  0]
 [ 0  0  2  0  0  0 40  0  8  0]
 [ 0  0  0  0  0  0  0 24  1 25]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  9 41]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.84      0.91        50
           1       0.00      0.00      0.00        50
           2       0.94      0.96      0.95        50
           3       1.00      0.58      0.73        50
           4       1.00      0.04      0.08        50
           5       0.58      0.14      0.23        50
           6       1.00      0.80      0.89        50
           7       1.00      0.48      0.65        50
           8       0.22      1.00      0.36        50
           9       0.56      0.82      0.67        50

    accuracy                           0.57       500
   macro avg       0.73      0.57      0.55       500
weighted avg       0.73      0.57      0.55       500

Generation time: 1.4483 seconds
Coverage=0.230
Avg set size=0.24
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.436    | 0.50     | 0.50    | 1.22     | 0.44    | 0.22    | 0.15    | 0      
1     | 50   | 0.131    | 0.00     | 0.00    | 2.70     | 0.33    | 0.14    | 0.12    | 8      
2     | 50   | 0.496    | 0.74     | 0.74    | 1.04     | 0.50    | 0.21    | 0.09    | 2      
3     | 50   | 0.293    | 0.00     | 0.00    | 1.50     | 0.32    | 0.26    | 0.17    | 3      
4     | 50   | 0.169    | 0.00     | 0.00    | 2.90     | 0.26    | 0.21    | 0.17    | 8      
5     | 50   | 0.297    | 0.08     | 0.02    | 1.86     | 0.36    | 0.29    | 0.09    | 8      
6     | 50   | 0.302    | 0.04     | 0.04    | 1.26     | 0.32    | 0.21    | 0.16    | 6      
7     | 50   | 0.291    | 0.04     | 0.04    | 1.62     | 0.32    | 0.27    | 0.16    | 9      
8     | 50   | 0.498    | 0.92     | 0.92    | 1.00     | 0.50    | 0.15    | 0.10    | 8      
9     | 50   | 0.330    | 0.04     | 0.04    | 1.18     | 0.34    | 0.24    | 0.12    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.6032 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.070    | 0.00     | 0.00    | 5.38     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.038    | 0.00     | 0.00    | 9.00     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.163    | 0.00     | 0.00    | 2.14     | 0.34    | 0.16    | 0.15    | 8      
3     | 50   | 0.117    | 0.00     | 0.00    | 3.50     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.04     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.184    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.10    | 8      
6     | 50   | 0.072    | 0.00     | 0.00    | 5.58     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 6.00     | 0.32    | 0.15    | 0.14    | 8      
8     | 50   | 0.374    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.141    | 0.00     | 0.00    | 2.78     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=8 | std=2.0 ---

== PCA [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.550
Confusion matrix (rows=true, cols=pred):
 [[19  1  3  3  1 15  0  2  5  1]
 [ 0 25  2  2  0  3  1  0 17  0]
 [ 0  0 34  5  3  3  1  1  3  0]
 [ 0  0 10 22  1  8  0  4  5  0]
 [ 0  0  3  0 28  7  2  2  5  3]
 [ 2  0  4  2  2 28  3  0  7  2]
 [ 0  0  3  3  4  7 29  0  4  0]
 [ 1  1  3  1  4  0  0 35  0  5]
 [ 1  1  3  6  0 12  0  0 25  2]
 [ 0  0  4  0  6  3  0  6  1 30]]
Classification report:
               precision    recall  f1-score   support

           0       0.83      0.38      0.52        50
           1       0.89      0.50      0.64        50
           2       0.49      0.68      0.57        50
           3       0.50      0.44      0.47        50
           4       0.57      0.56      0.57        50
           5       0.33      0.56      0.41        50
           6       0.81      0.58      0.67        50
           7       0.70      0.70      0.70        50
           8       0.35      0.50      0.41        50
           9       0.70      0.60      0.65        50

    accuracy                           0.55       500
   macro avg       0.62      0.55      0.56       500
weighted avg       0.62      0.55      0.56       500

Generation time: 0.5935 seconds
Coverage=0.464
Avg set size=0.76
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.319    | 0.76     | 0.32    | 2.44     | 0.56    | 0.20    | 0.10    | 0      
1     | 50   | 0.328    | 0.64     | 0.34    | 2.24     | 0.52    | 0.21    | 0.11    | 1      
2     | 50   | 0.535    | 0.86     | 0.62    | 1.90     | 0.68    | 0.16    | 0.07    | 2      
3     | 50   | 0.360    | 0.74     | 0.36    | 2.46     | 0.62    | 0.18    | 0.09    | 3      
4     | 50   | 0.429    | 0.74     | 0.44    | 1.94     | 0.56    | 0.20    | 0.10    | 4      
5     | 50   | 0.497    | 0.76     | 0.54    | 1.84     | 0.63    | 0.18    | 0.08    | 5      
6     | 50   | 0.415    | 0.78     | 0.50    | 1.94     | 0.56    | 0.22    | 0.10    | 6      
7     | 50   | 0.535    | 0.76     | 0.60    | 1.58     | 0.61    | 0.19    | 0.10    | 7      
8     | 50   | 0.376    | 0.78     | 0.40    | 2.08     | 0.60    | 0.19    | 0.09    | 8      
9     | 50   | 0.458    | 0.78     | 0.52    | 1.90     | 0.59    | 0.20    | 0.09    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.786
Confusion matrix (rows=true, cols=pred):
 [[47  0  0  0  0  1  0  0  2  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 47  0  0  0  0  0  3  0]
 [ 0  0  6 34  0  2  0  0  8  0]
 [ 1  0  0  0 24  0  0  0  9 16]
 [ 0  0  0  0  0 34  0  0 16  0]
 [ 0  0  9  0  0  0 37  0  4  0]
 [ 0  0  1  0  0  5  0 25  6 13]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  1  0  1  0  0  0  2 46]]
Classification report:
               precision    recall  f1-score   support

           0       0.98      0.94      0.96        50
           1       1.00      0.98      0.99        50
           2       0.73      0.94      0.82        50
           3       1.00      0.68      0.81        50
           4       0.96      0.48      0.64        50
           5       0.81      0.68      0.74        50
           6       1.00      0.74      0.85        50
           7       1.00      0.50      0.67        50
           8       0.50      1.00      0.66        50
           9       0.61      0.92      0.74        50

    accuracy                           0.79       500
   macro avg       0.86      0.79      0.79       500
weighted avg       0.86      0.79      0.79       500

Generation time: 0.9613 seconds
Coverage=0.484
Avg set size=0.50
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.527    | 0.70     | 0.70    | 1.08     | 0.53    | 0.20    | 0.12    | 0      
1     | 50   | 0.545    | 0.72     | 0.72    | 1.02     | 0.55    | 0.19    | 0.10    | 1      
2     | 50   | 0.473    | 0.62     | 0.62    | 1.06     | 0.48    | 0.20    | 0.10    | 2      
3     | 50   | 0.365    | 0.36     | 0.36    | 1.50     | 0.40    | 0.24    | 0.16    | 3      
4     | 50   | 0.294    | 0.16     | 0.14    | 1.84     | 0.33    | 0.24    | 0.15    | 4      
5     | 50   | 0.358    | 0.18     | 0.14    | 1.32     | 0.38    | 0.29    | 0.10    | 5      
6     | 50   | 0.366    | 0.34     | 0.32    | 1.46     | 0.39    | 0.18    | 0.14    | 6      
7     | 50   | 0.307    | 0.30     | 0.26    | 2.20     | 0.35    | 0.28    | 0.12    | 7      
8     | 50   | 0.563    | 0.84     | 0.84    | 1.00     | 0.56    | 0.14    | 0.10    | 8      
9     | 50   | 0.496    | 0.74     | 0.74    | 1.16     | 0.50    | 0.18    | 0.11    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.800
Confusion matrix (rows=true, cols=pred):
 [[46  0  0  0  0  2  0  0  2  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 48  0  0  0  0  0  2  0]
 [ 0  0  6 30  0  2  0  0 12  0]
 [ 0  0  0  0 33  2  0  0  8  7]
 [ 0  0  0  0  0 36  0  0 14  0]
 [ 0  0  6  0  0  3 36  0  5  0]
 [ 0  0  1  0  0  2  0 28  8 11]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  2  0  0  0  0  0  4 44]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.92      0.96        50
           1       1.00      0.98      0.99        50
           2       0.76      0.96      0.85        50
           3       1.00      0.60      0.75        50
           4       1.00      0.66      0.80        50
           5       0.77      0.72      0.74        50
           6       1.00      0.72      0.84        50
           7       1.00      0.56      0.72        50
           8       0.47      1.00      0.64        50
           9       0.71      0.88      0.79        50

    accuracy                           0.80       500
   macro avg       0.87      0.80      0.81       500
weighted avg       0.87      0.80      0.81       500

Generation time: 0.9524 seconds
Coverage=0.544
Avg set size=0.56
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.512    | 0.74     | 0.74    | 1.10     | 0.52    | 0.20    | 0.12    | 0      
1     | 50   | 0.697    | 0.96     | 0.96    | 1.02     | 0.70    | 0.13    | 0.07    | 1      
2     | 50   | 0.439    | 0.52     | 0.52    | 1.06     | 0.44    | 0.22    | 0.11    | 2      
3     | 50   | 0.347    | 0.30     | 0.28    | 1.70     | 0.39    | 0.24    | 0.16    | 3      
4     | 50   | 0.345    | 0.36     | 0.32    | 1.78     | 0.38    | 0.25    | 0.13    | 4      
5     | 50   | 0.371    | 0.28     | 0.26    | 1.28     | 0.38    | 0.29    | 0.10    | 5      
6     | 50   | 0.355    | 0.38     | 0.36    | 1.58     | 0.39    | 0.19    | 0.14    | 6      
7     | 50   | 0.323    | 0.38     | 0.34    | 2.10     | 0.36    | 0.27    | 0.12    | 7      
8     | 50   | 0.593    | 0.88     | 0.88    | 1.00     | 0.59    | 0.13    | 0.09    | 8      
9     | 50   | 0.517    | 0.78     | 0.78    | 1.30     | 0.53    | 0.17    | 0.10    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.666
Confusion matrix (rows=true, cols=pred):
 [[33  0  1  0  0 11  0  0  5  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 32  0  0  1  0  0 17  0]
 [ 0  0  2 27  0  5  0  0 16  0]
 [ 0  0  1  0 27  1  0  0 14  7]
 [ 0  0  0  0  0 31  0  0 19  0]
 [ 0  0  6  0  0  9 23  0 12  0]
 [ 0  0  1  0  0  1  0 19 18 11]
 [ 0  0  0  0  0  2  0  0 48  0]
 [ 0  0  0  0  0  0  0  0  6 44]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.66      0.80        50
           1       1.00      0.98      0.99        50
           2       0.74      0.64      0.69        50
           3       1.00      0.54      0.70        50
           4       1.00      0.54      0.70        50
           5       0.51      0.62      0.56        50
           6       1.00      0.46      0.63        50
           7       1.00      0.38      0.55        50
           8       0.31      0.96      0.47        50
           9       0.71      0.88      0.79        50

    accuracy                           0.67       500
   macro avg       0.83      0.67      0.69       500
weighted avg       0.83      0.67      0.69       500

Generation time: 0.9597 seconds
Coverage=0.344
Avg set size=0.35
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.329    | 0.26     | 0.26    | 1.68     | 0.36    | 0.23    | 0.16    | 0      
1     | 50   | 0.766    | 0.96     | 0.96    | 1.02     | 0.77    | 0.10    | 0.05    | 1      
2     | 50   | 0.298    | 0.00     | 0.00    | 1.40     | 0.32    | 0.26    | 0.14    | 2      
3     | 50   | 0.284    | 0.14     | 0.14    | 1.96     | 0.34    | 0.24    | 0.18    | 3      
4     | 50   | 0.296    | 0.26     | 0.24    | 2.04     | 0.35    | 0.24    | 0.14    | 4      
5     | 50   | 0.342    | 0.14     | 0.12    | 1.40     | 0.36    | 0.29    | 0.10    | 5      
6     | 50   | 0.262    | 0.20     | 0.20    | 2.42     | 0.32    | 0.20    | 0.16    | 6      
7     | 50   | 0.249    | 0.18     | 0.14    | 2.86     | 0.32    | 0.25    | 0.13    | 7      
8     | 50   | 0.552    | 0.78     | 0.78    | 1.04     | 0.55    | 0.15    | 0.10    | 8      
9     | 50   | 0.461    | 0.60     | 0.60    | 1.20     | 0.47    | 0.18    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.8414 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.089    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.043    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.049    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.042    | 0.00     | 0.00    | 7.64     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.916
Confusion matrix (rows=true, cols=pred):
 [[42  0  1  0  0  4  1  0  1  1]
 [ 0 47  0  0  0  1  0  0  2  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  1 48  0  0  0  0  1  0]
 [ 0  0  0  0 37  0  1  0  0 12]
 [ 0  1  0  2  0 44  0  0  3  0]
 [ 0  0  0  0  0  0 49  0  1  0]
 [ 0  0  0  0  0  0  0 44  0  6]
 [ 0  0  2  0  0  0  0  0 48  0]
 [ 0  0  0  0  0  0  0  1  0 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.84      0.91        50
           1       0.98      0.94      0.96        50
           2       0.93      1.00      0.96        50
           3       0.96      0.96      0.96        50
           4       1.00      0.74      0.85        50
           5       0.90      0.88      0.89        50
           6       0.96      0.98      0.97        50
           7       0.98      0.88      0.93        50
           8       0.86      0.96      0.91        50
           9       0.72      0.98      0.83        50

    accuracy                           0.92       500
   macro avg       0.93      0.92      0.92       500
weighted avg       0.93      0.92      0.92       500

Generation time: 0.8674 seconds
Coverage=0.836
Avg set size=0.88
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.730    | 0.84     | 0.80    | 1.54     | 0.76    | 0.12    | 0.05    | 0      
1     | 50   | 0.483    | 0.66     | 0.66    | 1.06     | 0.49    | 0.18    | 0.12    | 1      
2     | 50   | 0.886    | 0.98     | 0.98    | 1.00     | 0.89    | 0.06    | 0.02    | 2      
3     | 50   | 0.712    | 0.88     | 0.86    | 1.08     | 0.72    | 0.13    | 0.07    | 3      
4     | 50   | 0.536    | 0.78     | 0.68    | 1.30     | 0.57    | 0.21    | 0.08    | 4      
5     | 50   | 0.551    | 0.80     | 0.74    | 1.12     | 0.57    | 0.24    | 0.08    | 5      
6     | 50   | 0.835    | 0.94     | 0.94    | 1.04     | 0.84    | 0.08    | 0.04    | 6      
7     | 50   | 0.769    | 1.02     | 0.90    | 1.12     | 0.79    | 0.15    | 0.03    | 7      
8     | 50   | 0.676    | 0.96     | 0.92    | 1.08     | 0.69    | 0.13    | 0.07    | 8      
9     | 50   | 0.637    | 0.92     | 0.88    | 1.02     | 0.64    | 0.18    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=3.58354e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0653 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=4.30212e-09): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.522
Confusion matrix (rows=true, cols=pred):
 [[31  0  5  1  0  9  0  0  4  0]
 [ 0 35  0  0  0  0  0  0 15  0]
 [ 1  0 32  3  1  6  5  0  2  0]
 [ 0  1  3 27  0  6  7  0  5  1]
 [ 0  0  0  0  3  1  2  0  3 41]
 [ 0  0 12  3  0 15  0  3 16  1]
 [ 1  0  7  5  0  7 23  0  7  0]
 [ 0  1  0  5  0  0  0 26  5 13]
 [ 0  0  3 11  0  3  0  0 31  2]
 [ 0  0  1  5  0  0  0  4  2 38]]
Classification report:
               precision    recall  f1-score   support

           0       0.94      0.62      0.75        50
           1       0.95      0.70      0.80        50
           2       0.51      0.64      0.57        50
           3       0.45      0.54      0.49        50
           4       0.75      0.06      0.11        50
           5       0.32      0.30      0.31        50
           6       0.62      0.46      0.53        50
           7       0.79      0.52      0.63        50
           8       0.34      0.62      0.44        50
           9       0.40      0.76      0.52        50

    accuracy                           0.52       500
   macro avg       0.61      0.52      0.51       500
weighted avg       0.61      0.52      0.51       500

Generation time: 1.0646 seconds
Coverage=0.430
Avg set size=0.67
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.579    | 0.74     | 0.60    | 2.48     | 0.69    | 0.14    | 0.06    | 0      
1     | 50   | 0.379    | 0.38     | 0.38    | 1.32     | 0.41    | 0.23    | 0.11    | 1      
2     | 50   | 0.529    | 0.84     | 0.58    | 2.14     | 0.72    | 0.14    | 0.06    | 2      
3     | 50   | 0.494    | 0.80     | 0.52    | 3.04     | 0.67    | 0.16    | 0.07    | 3      
4     | 50   | 0.162    | 0.40     | 0.00    | 3.12     | 0.40    | 0.22    | 0.13    | 9      
5     | 50   | 0.266    | 0.46     | 0.24    | 2.86     | 0.45    | 0.22    | 0.12    | 8      
6     | 50   | 0.414    | 0.78     | 0.42    | 3.50     | 0.64    | 0.18    | 0.09    | 6      
7     | 50   | 0.491    | 0.88     | 0.50    | 2.08     | 0.67    | 0.20    | 0.06    | 7      
8     | 50   | 0.376    | 0.54     | 0.38    | 1.50     | 0.45    | 0.23    | 0.12    | 8      
9     | 50   | 0.543    | 0.84     | 0.68    | 1.50     | 0.61    | 0.18    | 0.09    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.554
Confusion matrix (rows=true, cols=pred):
 [[39  0  1  0  0  7  0  0  3  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 44  0  0  0  0  0  6  0]
 [ 0  0  1 29  0  0  0  0 20  0]
 [ 0  0  0  0  3  0  0  0 38  9]
 [ 0  0  0  0  0 10  0  0 40  0]
 [ 0  0  3  0  0  0 39  0  8  0]
 [ 0  0  0  1  0  0  0 24  5 20]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 11 39]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.78      0.88        50
           1       0.00      0.00      0.00        50
           2       0.90      0.88      0.89        50
           3       0.97      0.58      0.72        50
           4       1.00      0.06      0.11        50
           5       0.59      0.20      0.30        50
           6       1.00      0.78      0.88        50
           7       1.00      0.48      0.65        50
           8       0.22      1.00      0.36        50
           9       0.57      0.78      0.66        50

    accuracy                           0.55       500
   macro avg       0.72      0.55      0.54       500
weighted avg       0.72      0.55      0.54       500

Generation time: 0.9515 seconds
Coverage=0.248
Avg set size=0.26
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.445    | 0.54     | 0.52    | 1.38     | 0.47    | 0.21    | 0.14    | 0      
1     | 50   | 0.130    | 0.00     | 0.00    | 2.80     | 0.33    | 0.14    | 0.12    | 8      
2     | 50   | 0.491    | 0.68     | 0.68    | 1.12     | 0.50    | 0.20    | 0.10    | 2      
3     | 50   | 0.295    | 0.00     | 0.00    | 1.56     | 0.33    | 0.25    | 0.17    | 3      
4     | 50   | 0.169    | 0.00     | 0.00    | 2.86     | 0.27    | 0.21    | 0.16    | 8      
5     | 50   | 0.293    | 0.16     | 0.04    | 1.80     | 0.36    | 0.28    | 0.10    | 8      
6     | 50   | 0.304    | 0.20     | 0.18    | 1.30     | 0.33    | 0.21    | 0.15    | 6      
7     | 50   | 0.296    | 0.12     | 0.12    | 1.68     | 0.33    | 0.26    | 0.15    | 7      
8     | 50   | 0.489    | 0.86     | 0.86    | 1.00     | 0.49    | 0.16    | 0.10    | 8      
9     | 50   | 0.330    | 0.08     | 0.08    | 1.22     | 0.34    | 0.23    | 0.13    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=8² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9709 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.071    | 0.00     | 0.00    | 5.18     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.038    | 0.00     | 0.00    | 8.94     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.163    | 0.00     | 0.00    | 2.24     | 0.34    | 0.16    | 0.15    | 8      
3     | 50   | 0.117    | 0.00     | 0.00    | 3.48     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.06     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.184    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.10    | 8      
6     | 50   | 0.072    | 0.00     | 0.00    | 5.56     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.067    | 0.00     | 0.00    | 5.94     | 0.32    | 0.15    | 0.14    | 8      
8     | 50   | 0.374    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.141    | 0.00     | 0.00    | 2.78     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=10 | std=1.0 ---

== PCA [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.810
Confusion matrix (rows=true, cols=pred):
 [[43  0  1  0  0  5  1  0  0  0]
 [ 0 35  0  0  0  0  0  0 15  0]
 [ 0  0 31  1  0  5  0  2 10  1]
 [ 0  0  5 28  0 10  0  0  5  2]
 [ 0  0  1  0 46  1  0  0  0  2]
 [ 0  0  1  1  1 39  2  0  6  0]
 [ 0  0  1  0  0  0 48  0  1  0]
 [ 0  0  0  0  0  0  0 48  0  2]
 [ 0  0  1  2  1  1  0  0 43  2]
 [ 0  0  0  0  0  1  0  4  1 44]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.86      0.92        50
           1       1.00      0.70      0.82        50
           2       0.76      0.62      0.68        50
           3       0.88      0.56      0.68        50
           4       0.96      0.92      0.94        50
           5       0.63      0.78      0.70        50
           6       0.94      0.96      0.95        50
           7       0.89      0.96      0.92        50
           8       0.53      0.86      0.66        50
           9       0.83      0.88      0.85        50

    accuracy                           0.81       500
   macro avg       0.84      0.81      0.81       500
weighted avg       0.84      0.81      0.81       500

Generation time: 0.8042 seconds
Coverage=0.644
Avg set size=0.72
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.469    | 0.74     | 0.62    | 1.16     | 0.50    | 0.22    | 0.09    | 0      
1     | 50   | 0.446    | 0.60     | 0.56    | 1.32     | 0.48    | 0.24    | 0.11    | 1      
2     | 50   | 0.415    | 0.60     | 0.46    | 1.80     | 0.49    | 0.18    | 0.11    | 2      
3     | 50   | 0.380    | 0.50     | 0.36    | 1.90     | 0.48    | 0.20    | 0.12    | 3      
4     | 50   | 0.595    | 0.80     | 0.78    | 1.10     | 0.60    | 0.16    | 0.10    | 4      
5     | 50   | 0.527    | 0.74     | 0.68    | 1.28     | 0.55    | 0.21    | 0.08    | 5      
6     | 50   | 0.575    | 0.82     | 0.80    | 1.04     | 0.58    | 0.17    | 0.09    | 6      
7     | 50   | 0.624    | 0.96     | 0.88    | 1.04     | 0.63    | 0.19    | 0.07    | 7      
8     | 50   | 0.499    | 0.70     | 0.66    | 1.18     | 0.52    | 0.18    | 0.10    | 8      
9     | 50   | 0.488    | 0.72     | 0.64    | 1.18     | 0.51    | 0.20    | 0.10    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.872
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  1 35  0  4  0  0 10  0]
 [ 0  0  0  0 31  0  0  0  3 16]
 [ 0  0  0  0  0 38  0  0 12  0]
 [ 0  0  4  0  0  0 44  0  2  0]
 [ 0  0  0  0  0  0  0 38  5  7]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       0.91      1.00      0.95        50
           3       1.00      0.70      0.82        50
           4       1.00      0.62      0.77        50
           5       0.90      0.76      0.83        50
           6       1.00      0.88      0.94        50
           7       1.00      0.76      0.86        50
           8       0.61      1.00      0.76        50
           9       0.68      1.00      0.81        50

    accuracy                           0.87       500
   macro avg       0.91      0.87      0.87       500
weighted avg       0.91      0.87      0.87       500

Generation time: 1.0095 seconds
Coverage=0.524
Avg set size=0.52
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.543    | 0.84     | 0.84    | 1.00     | 0.54    | 0.20    | 0.11    | 0      
1     | 50   | 0.580    | 0.92     | 0.92    | 1.00     | 0.58    | 0.17    | 0.08    | 1      
2     | 50   | 0.492    | 0.88     | 0.88    | 1.00     | 0.49    | 0.20    | 0.09    | 2      
3     | 50   | 0.335    | 0.20     | 0.20    | 1.46     | 0.36    | 0.25    | 0.18    | 3      
4     | 50   | 0.311    | 0.06     | 0.06    | 1.42     | 0.33    | 0.26    | 0.17    | 4      
5     | 50   | 0.378    | 0.10     | 0.10    | 1.24     | 0.39    | 0.30    | 0.09    | 5      
6     | 50   | 0.379    | 0.42     | 0.42    | 1.16     | 0.39    | 0.18    | 0.14    | 6      
7     | 50   | 0.309    | 0.06     | 0.06    | 1.42     | 0.32    | 0.26    | 0.14    | 7      
8     | 50   | 0.591    | 0.98     | 0.98    | 1.00     | 0.59    | 0.13    | 0.09    | 8      
9     | 50   | 0.509    | 0.78     | 0.78    | 1.00     | 0.51    | 0.16    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.876
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 29  0  4  0  0 17  0]
 [ 0  0  0  0 38  0  0  0  5  7]
 [ 0  0  0  0  0 42  0  0  8  0]
 [ 0  0  3  0  0  1 44  0  2  0]
 [ 0  0  0  0  0  0  0 37  6  7]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  2 48]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       0.94      1.00      0.97        50
           3       1.00      0.58      0.73        50
           4       1.00      0.76      0.86        50
           5       0.89      0.84      0.87        50
           6       1.00      0.88      0.94        50
           7       1.00      0.74      0.85        50
           8       0.56      1.00      0.71        50
           9       0.77      0.96      0.86        50

    accuracy                           0.88       500
   macro avg       0.92      0.88      0.88       500
weighted avg       0.92      0.88      0.88       500

Generation time: 0.9850 seconds
Coverage=0.550
Avg set size=0.55
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.523    | 0.84     | 0.84    | 1.00     | 0.52    | 0.21    | 0.11    | 0      
1     | 50   | 0.741    | 1.00     | 1.00    | 1.00     | 0.74    | 0.10    | 0.05    | 1      
2     | 50   | 0.449    | 0.62     | 0.62    | 1.00     | 0.45    | 0.21    | 0.10    | 2      
3     | 50   | 0.320    | 0.16     | 0.16    | 1.66     | 0.35    | 0.24    | 0.18    | 3      
4     | 50   | 0.353    | 0.22     | 0.22    | 1.26     | 0.36    | 0.27    | 0.15    | 4      
5     | 50   | 0.389    | 0.28     | 0.28    | 1.16     | 0.39    | 0.30    | 0.09    | 5      
6     | 50   | 0.408    | 0.52     | 0.52    | 1.20     | 0.41    | 0.17    | 0.14    | 6      
7     | 50   | 0.300    | 0.08     | 0.08    | 1.46     | 0.31    | 0.25    | 0.14    | 7      
8     | 50   | 0.617    | 1.00     | 1.00    | 1.00     | 0.62    | 0.12    | 0.08    | 8      
9     | 50   | 0.529    | 0.78     | 0.78    | 1.04     | 0.53    | 0.16    | 0.12    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.734
Confusion matrix (rows=true, cols=pred):
 [[46  0  0  0  0  4  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 39  0  0  0  0  0 11  0]
 [ 0  0  0 17  0  2  0  0 31  0]
 [ 0  0  0  0 31  0  0  0 10  9]
 [ 0  0  0  0  0 35  0  0 15  0]
 [ 0  0  2  0  0  2 39  0  7  0]
 [ 0  0  0  0  0  2  0 15 19 14]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  5 45]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.92      0.96        50
           1       1.00      1.00      1.00        50
           2       0.95      0.78      0.86        50
           3       1.00      0.34      0.51        50
           4       1.00      0.62      0.77        50
           5       0.78      0.70      0.74        50
           6       1.00      0.78      0.88        50
           7       1.00      0.30      0.46        50
           8       0.34      1.00      0.51        50
           9       0.66      0.90      0.76        50

    accuracy                           0.73       500
   macro avg       0.87      0.73      0.74       500
weighted avg       0.87      0.73      0.74       500

Generation time: 1.6480 seconds
Coverage=0.326
Avg set size=0.33
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.376    | 0.30     | 0.30    | 1.08     | 0.38    | 0.23    | 0.16    | 0      
1     | 50   | 0.806    | 1.00     | 1.00    | 1.00     | 0.81    | 0.07    | 0.04    | 1      
2     | 50   | 0.308    | 0.00     | 0.00    | 1.24     | 0.32    | 0.26    | 0.14    | 2      
3     | 50   | 0.255    | 0.00     | 0.00    | 1.98     | 0.32    | 0.25    | 0.19    | 8      
4     | 50   | 0.315    | 0.16     | 0.16    | 1.52     | 0.33    | 0.25    | 0.16    | 4      
5     | 50   | 0.343    | 0.00     | 0.00    | 1.30     | 0.35    | 0.30    | 0.09    | 5      
6     | 50   | 0.300    | 0.10     | 0.10    | 1.48     | 0.31    | 0.19    | 0.17    | 6      
7     | 50   | 0.214    | 0.02     | 0.02    | 2.64     | 0.27    | 0.23    | 0.15    | 8      
8     | 50   | 0.587    | 1.00     | 1.00    | 1.00     | 0.59    | 0.13    | 0.09    | 8      
9     | 50   | 0.476    | 0.68     | 0.68    | 1.10     | 0.48    | 0.17    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.8938 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 7.92     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.992
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 50  0  0  0  0  0  0]
 [ 0  0  0  0 48  0  0  0  0  2]
 [ 0  0  0  0  0 48  0  0  2  0]
 [ 0  0  0  0  0  0 50  0  0  0]
 [ 0  0  0  0  0  0  0 50  0  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       1.00      1.00      1.00        50
           3       1.00      1.00      1.00        50
           4       1.00      0.96      0.98        50
           5       1.00      0.96      0.98        50
           6       1.00      1.00      1.00        50
           7       1.00      1.00      1.00        50
           8       0.96      1.00      0.98        50
           9       0.96      1.00      0.98        50

    accuracy                           0.99       500
   macro avg       0.99      0.99      0.99       500
weighted avg       0.99      0.99      0.99       500

Generation time: 0.8585 seconds
Coverage=0.974
Avg set size=0.98
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.833    | 0.96     | 0.96    | 1.00     | 0.83    | 0.08    | 0.04    | 0      
1     | 50   | 0.574    | 0.98     | 0.98    | 1.00     | 0.57    | 0.17    | 0.09    | 1      
2     | 50   | 0.894    | 0.98     | 0.98    | 1.00     | 0.89    | 0.05    | 0.03    | 2      
3     | 50   | 0.819    | 1.00     | 1.00    | 1.00     | 0.82    | 0.08    | 0.05    | 3      
4     | 50   | 0.585    | 0.92     | 0.92    | 1.04     | 0.59    | 0.23    | 0.07    | 4      
5     | 50   | 0.616    | 0.94     | 0.92    | 1.04     | 0.62    | 0.24    | 0.06    | 5      
6     | 50   | 0.853    | 0.98     | 0.98    | 1.00     | 0.85    | 0.06    | 0.03    | 6      
7     | 50   | 0.831    | 1.00     | 1.00    | 1.00     | 0.83    | 0.13    | 0.02    | 7      
8     | 50   | 0.719    | 1.00     | 1.00    | 1.00     | 0.72    | 0.12    | 0.06    | 8      
9     | 50   | 0.694    | 1.00     | 1.00    | 1.00     | 0.69    | 0.13    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=3.60332e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.1062 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=7.83564e-09): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.940
Confusion matrix (rows=true, cols=pred):
 [[38  0  0  0  0 10  1  0  1  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 49  0  0  0  0  1  0]
 [ 0  0  0  0 46  0  0  0  0  4]
 [ 0  0  0  1  1 41  0  0  7  0]
 [ 0  0  0  0  0  1 49  0  0  0]
 [ 0  0  0  0  0  0  0 47  0  3]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.76      0.86        50
           1       1.00      1.00      1.00        50
           2       1.00      1.00      1.00        50
           3       0.98      0.98      0.98        50
           4       0.98      0.92      0.95        50
           5       0.79      0.82      0.80        50
           6       0.98      0.98      0.98        50
           7       1.00      0.94      0.97        50
           8       0.85      1.00      0.92        50
           9       0.88      1.00      0.93        50

    accuracy                           0.94       500
   macro avg       0.95      0.94      0.94       500
weighted avg       0.95      0.94      0.94       500

Generation time: 1.0818 seconds
Coverage=0.918
Avg set size=0.95
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.688    | 0.82     | 0.70    | 1.48     | 0.75    | 0.13    | 0.06    | 0      
1     | 50   | 0.706    | 1.00     | 1.00    | 1.00     | 0.71    | 0.10    | 0.07    | 1      
2     | 50   | 0.876    | 0.98     | 0.98    | 1.00     | 0.88    | 0.06    | 0.02    | 2      
3     | 50   | 0.825    | 0.94     | 0.94    | 1.04     | 0.83    | 0.09    | 0.05    | 3      
4     | 50   | 0.597    | 0.92     | 0.86    | 1.08     | 0.60    | 0.24    | 0.07    | 4      
5     | 50   | 0.551    | 0.88     | 0.80    | 1.20     | 0.57    | 0.25    | 0.07    | 5      
6     | 50   | 0.902    | 0.98     | 0.98    | 1.02     | 0.90    | 0.04    | 0.02    | 6      
7     | 50   | 0.800    | 1.00     | 0.94    | 1.06     | 0.82    | 0.15    | 0.02    | 7      
8     | 50   | 0.694    | 0.98     | 0.98    | 1.00     | 0.69    | 0.15    | 0.06    | 8      
9     | 50   | 0.717    | 1.00     | 1.00    | 1.00     | 0.72    | 0.16    | 0.07    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.562
Confusion matrix (rows=true, cols=pred):
 [[39  0  0  0  0 10  0  0  1  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 27  0  0  0  0 23  0]
 [ 0  0  0  0  0  0  0  0 39 11]
 [ 0  0  0  0  0 11  0  0 39  0]
 [ 0  0  2  0  0  0 36  0 12  0]
 [ 0  0  0  0  0  0  0 20  0 30]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  2 48]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.78      0.88        50
           1       0.00      0.00      0.00        50
           2       0.96      1.00      0.98        50
           3       1.00      0.54      0.70        50
           4       0.00      0.00      0.00        50
           5       0.52      0.22      0.31        50
           6       1.00      0.72      0.84        50
           7       1.00      0.40      0.57        50
           8       0.23      1.00      0.38        50
           9       0.54      0.96      0.69        50

    accuracy                           0.56       500
   macro avg       0.63      0.56      0.53       500
weighted avg       0.63      0.56      0.53       500

Generation time: 1.0001 seconds
Coverage=0.220
Avg set size=0.22
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.402    | 0.44     | 0.44    | 1.32     | 0.42    | 0.23    | 0.15    | 0      
1     | 50   | 0.131    | 0.00     | 0.00    | 2.56     | 0.33    | 0.14    | 0.12    | 8      
2     | 50   | 0.466    | 0.72     | 0.72    | 1.00     | 0.47    | 0.22    | 0.09    | 2      
3     | 50   | 0.295    | 0.00     | 0.00    | 1.46     | 0.33    | 0.26    | 0.17    | 3      
4     | 50   | 0.166    | 0.00     | 0.00    | 2.98     | 0.26    | 0.22    | 0.17    | 8      
5     | 50   | 0.307    | 0.02     | 0.02    | 1.78     | 0.36    | 0.30    | 0.09    | 8      
6     | 50   | 0.284    | 0.00     | 0.00    | 1.32     | 0.30    | 0.20    | 0.17    | 6      
7     | 50   | 0.281    | 0.04     | 0.04    | 1.78     | 0.31    | 0.26    | 0.17    | 9      
8     | 50   | 0.499    | 0.94     | 0.94    | 1.00     | 0.50    | 0.15    | 0.10    | 8      
9     | 50   | 0.343    | 0.04     | 0.04    | 1.04     | 0.34    | 0.24    | 0.11    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9784 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.068    | 0.00     | 0.00    | 5.54     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.038    | 0.00     | 0.00    | 8.98     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.159    | 0.00     | 0.00    | 2.24     | 0.34    | 0.16    | 0.15    | 8      
3     | 50   | 0.116    | 0.00     | 0.00    | 3.58     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.00     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.185    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.10    | 8      
6     | 50   | 0.071    | 0.00     | 0.00    | 5.66     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.065    | 0.00     | 0.00    | 6.00     | 0.33    | 0.15    | 0.14    | 8      
8     | 50   | 0.373    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.142    | 0.00     | 0.00    | 2.72     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=10 | std=1.5 ---

== PCA [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.664
Confusion matrix (rows=true, cols=pred):
 [[33  0  4  1  0  9  1  0  2  0]
 [ 0 28  0  2  0  2  0  0 18  0]
 [ 0  1 28  1  1  8  1  3  7  0]
 [ 0  0  8 23  1 11  0  1  3  3]
 [ 0  0  1  0 39  6  0  0  1  3]
 [ 0  0  2  2  2 34  2  0  7  1]
 [ 1  0  2  0  2  4 38  0  3  0]
 [ 0  0  1  0  1  0  0 44  0  4]
 [ 0  1  1  3  4  7  1  0 30  3]
 [ 2  0  0  0  0  5  0  7  1 35]]
Classification report:
               precision    recall  f1-score   support

           0       0.92      0.66      0.77        50
           1       0.93      0.56      0.70        50
           2       0.60      0.56      0.58        50
           3       0.72      0.46      0.56        50
           4       0.78      0.78      0.78        50
           5       0.40      0.68      0.50        50
           6       0.88      0.76      0.82        50
           7       0.80      0.88      0.84        50
           8       0.42      0.60      0.49        50
           9       0.71      0.70      0.71        50

    accuracy                           0.66       500
   macro avg       0.72      0.66      0.67       500
weighted avg       0.72      0.66      0.67       500

Generation time: 0.7266 seconds
Coverage=0.558
Avg set size=0.71
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.412    | 0.64     | 0.46    | 1.42     | 0.50    | 0.23    | 0.10    | 0      
1     | 50   | 0.432    | 0.66     | 0.50    | 1.60     | 0.50    | 0.22    | 0.10    | 1      
2     | 50   | 0.427    | 0.72     | 0.48    | 2.16     | 0.57    | 0.18    | 0.10    | 2      
3     | 50   | 0.389    | 0.62     | 0.38    | 2.26     | 0.55    | 0.19    | 0.11    | 3      
4     | 50   | 0.552    | 0.80     | 0.70    | 1.40     | 0.60    | 0.16    | 0.10    | 4      
5     | 50   | 0.503    | 0.78     | 0.60    | 1.58     | 0.59    | 0.21    | 0.09    | 5      
6     | 50   | 0.534    | 0.80     | 0.70    | 1.32     | 0.57    | 0.20    | 0.10    | 6      
7     | 50   | 0.619    | 0.90     | 0.80    | 1.12     | 0.64    | 0.18    | 0.07    | 7      
8     | 50   | 0.421    | 0.62     | 0.48    | 1.68     | 0.51    | 0.21    | 0.11    | 8      
9     | 50   | 0.423    | 0.60     | 0.48    | 1.48     | 0.50    | 0.22    | 0.11    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.808
Confusion matrix (rows=true, cols=pred):
 [[48  0  0  0  0  1  0  0  1  0]
 [ 0 48  0  0  0  0  0  0  2  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  4 31  0  5  0  0 10  0]
 [ 0  0  0  0 26  0  0  0  6 18]
 [ 0  0  0  0  0 36  0  0 14  0]
 [ 0  0  9  0  0  2 37  0  2  0]
 [ 0  0  2  0  0  1  0 32  5 10]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  4 46]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.96      0.98        50
           1       1.00      0.96      0.98        50
           2       0.77      1.00      0.87        50
           3       1.00      0.62      0.77        50
           4       1.00      0.52      0.68        50
           5       0.80      0.72      0.76        50
           6       1.00      0.74      0.85        50
           7       1.00      0.64      0.78        50
           8       0.53      1.00      0.69        50
           9       0.62      0.92      0.74        50

    accuracy                           0.81       500
   macro avg       0.87      0.81      0.81       500
weighted avg       0.87      0.81      0.81       500

Generation time: 1.6476 seconds
Coverage=0.522
Avg set size=0.53
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.536    | 0.76     | 0.76    | 1.06     | 0.54    | 0.20    | 0.11    | 0      
1     | 50   | 0.567    | 0.82     | 0.82    | 1.04     | 0.57    | 0.17    | 0.09    | 1      
2     | 50   | 0.486    | 0.78     | 0.78    | 1.00     | 0.49    | 0.20    | 0.09    | 2      
3     | 50   | 0.337    | 0.34     | 0.34    | 1.80     | 0.38    | 0.24    | 0.17    | 3      
4     | 50   | 0.309    | 0.16     | 0.12    | 1.58     | 0.35    | 0.25    | 0.16    | 4      
5     | 50   | 0.368    | 0.20     | 0.18    | 1.28     | 0.39    | 0.29    | 0.10    | 5      
6     | 50   | 0.372    | 0.46     | 0.44    | 1.44     | 0.40    | 0.18    | 0.14    | 6      
7     | 50   | 0.303    | 0.16     | 0.14    | 1.70     | 0.33    | 0.26    | 0.13    | 7      
8     | 50   | 0.560    | 0.90     | 0.90    | 1.00     | 0.56    | 0.15    | 0.09    | 8      
9     | 50   | 0.508    | 0.74     | 0.74    | 1.08     | 0.51    | 0.16    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.838
Confusion matrix (rows=true, cols=pred):
 [[49  0  0  0  0  1  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  3 27  0  5  0  0 15  0]
 [ 0  0  0  0 34  0  0  0  6 10]
 [ 0  0  0  0  0 39  0  0 11  0]
 [ 0  0  6  0  0  1 41  0  2  0]
 [ 0  0  1  0  0  1  0 34  8  6]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  1  0  0  1  0  0  2 46]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.98      0.99        50
           1       1.00      1.00      1.00        50
           2       0.82      0.98      0.89        50
           3       1.00      0.54      0.70        50
           4       1.00      0.68      0.81        50
           5       0.81      0.78      0.80        50
           6       1.00      0.82      0.90        50
           7       1.00      0.68      0.81        50
           8       0.53      1.00      0.69        50
           9       0.74      0.92      0.82        50

    accuracy                           0.84       500
   macro avg       0.89      0.84      0.84       500
weighted avg       0.89      0.84      0.84       500

Generation time: 0.9820 seconds
Coverage=0.540
Avg set size=0.55
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.517    | 0.78     | 0.78    | 1.02     | 0.52    | 0.20    | 0.11    | 0      
1     | 50   | 0.720    | 0.96     | 0.96    | 1.00     | 0.72    | 0.11    | 0.06    | 1      
2     | 50   | 0.447    | 0.60     | 0.60    | 1.02     | 0.45    | 0.21    | 0.10    | 2      
3     | 50   | 0.319    | 0.30     | 0.26    | 2.00     | 0.37    | 0.23    | 0.17    | 3      
4     | 50   | 0.352    | 0.26     | 0.24    | 1.50     | 0.38    | 0.26    | 0.14    | 4      
5     | 50   | 0.379    | 0.30     | 0.30    | 1.22     | 0.39    | 0.28    | 0.10    | 5      
6     | 50   | 0.410    | 0.52     | 0.52    | 1.40     | 0.43    | 0.18    | 0.13    | 6      
7     | 50   | 0.299    | 0.14     | 0.14    | 1.82     | 0.33    | 0.25    | 0.13    | 7      
8     | 50   | 0.587    | 0.90     | 0.90    | 1.00     | 0.59    | 0.14    | 0.09    | 8      
9     | 50   | 0.522    | 0.70     | 0.70    | 1.18     | 0.53    | 0.16    | 0.11    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.700
Confusion matrix (rows=true, cols=pred):
 [[44  0  0  0  0  4  0  0  2  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 35  0  0  3  0  0 12  0]
 [ 0  0  0 20  0  5  0  0 25  0]
 [ 0  0  0  0 28  0  0  0 12 10]
 [ 0  0  0  0  0 32  0  0 18  0]
 [ 0  0  4  0  0  3 36  0  7  0]
 [ 0  0  0  0  0  4  0 16 17 13]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 10 40]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.88      0.94        50
           1       1.00      0.98      0.99        50
           2       0.90      0.70      0.79        50
           3       1.00      0.40      0.57        50
           4       1.00      0.56      0.72        50
           5       0.63      0.64      0.63        50
           6       1.00      0.72      0.84        50
           7       1.00      0.32      0.48        50
           8       0.32      1.00      0.49        50
           9       0.63      0.80      0.71        50

    accuracy                           0.70       500
   macro avg       0.85      0.70      0.72       500
weighted avg       0.85      0.70      0.72       500

Generation time: 1.0040 seconds
Coverage=0.362
Avg set size=0.36
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.383    | 0.34     | 0.34    | 1.22     | 0.39    | 0.23    | 0.15    | 0      
1     | 50   | 0.774    | 0.94     | 0.94    | 1.02     | 0.77    | 0.08    | 0.05    | 1      
2     | 50   | 0.312    | 0.06     | 0.06    | 1.36     | 0.33    | 0.25    | 0.14    | 2      
3     | 50   | 0.260    | 0.10     | 0.10    | 2.08     | 0.34    | 0.24    | 0.18    | 8      
4     | 50   | 0.321    | 0.26     | 0.26    | 1.84     | 0.36    | 0.25    | 0.14    | 4      
5     | 50   | 0.336    | 0.06     | 0.06    | 1.36     | 0.35    | 0.29    | 0.10    | 5      
6     | 50   | 0.313    | 0.24     | 0.24    | 1.70     | 0.34    | 0.19    | 0.16    | 6      
7     | 50   | 0.219    | 0.06     | 0.06    | 2.92     | 0.29    | 0.23    | 0.14    | 8      
8     | 50   | 0.566    | 0.92     | 0.92    | 1.00     | 0.57    | 0.14    | 0.10    | 8      
9     | 50   | 0.471    | 0.64     | 0.64    | 1.42     | 0.49    | 0.16    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9021 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.049    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.042    | 0.00     | 0.00    | 7.72     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.966
Confusion matrix (rows=true, cols=pred):
 [[46  0  1  0  0  0  0  0  3  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  0 50  0  0  0  0  0  0]
 [ 0  0  0  0 46  0  0  0  0  4]
 [ 0  0  0  1  1 44  0  0  4  0]
 [ 0  0  0  0  0  0 49  0  1  0]
 [ 0  0  0  0  0  0  0 50  0  0]
 [ 0  0  0  1  0  0  0  0 49  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.92      0.96        50
           1       1.00      1.00      1.00        50
           2       0.98      0.98      0.98        50
           3       0.96      1.00      0.98        50
           4       0.98      0.92      0.95        50
           5       1.00      0.88      0.94        50
           6       1.00      0.98      0.99        50
           7       1.00      1.00      1.00        50
           8       0.84      0.98      0.91        50
           9       0.93      1.00      0.96        50

    accuracy                           0.97       500
   macro avg       0.97      0.97      0.97       500
weighted avg       0.97      0.97      0.97       500

Generation time: 1.3546 seconds
Coverage=0.912
Avg set size=0.93
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.753    | 0.86     | 0.86    | 1.16     | 0.77    | 0.11    | 0.05    | 0      
1     | 50   | 0.563    | 0.84     | 0.84    | 1.00     | 0.56    | 0.17    | 0.09    | 1      
2     | 50   | 0.855    | 0.98     | 0.98    | 1.04     | 0.86    | 0.07    | 0.03    | 2      
3     | 50   | 0.815    | 0.96     | 0.96    | 1.00     | 0.81    | 0.09    | 0.04    | 3      
4     | 50   | 0.568    | 0.86     | 0.80    | 1.08     | 0.58    | 0.21    | 0.08    | 4      
5     | 50   | 0.597    | 0.98     | 0.88    | 1.14     | 0.62    | 0.21    | 0.08    | 5      
6     | 50   | 0.815    | 0.96     | 0.96    | 1.06     | 0.82    | 0.08    | 0.04    | 6      
7     | 50   | 0.826    | 1.00     | 1.00    | 1.00     | 0.83    | 0.12    | 0.02    | 7      
8     | 50   | 0.662    | 0.92     | 0.90    | 1.02     | 0.67    | 0.15    | 0.07    | 8      
9     | 50   | 0.686    | 0.94     | 0.94    | 1.00     | 0.69    | 0.14    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=3.60332e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0984 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=7.83564e-09): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.866
Confusion matrix (rows=true, cols=pred):
 [[33  0  1  1  1  9  1  0  3  1]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 45  0  0  1  2  0  2  0]
 [ 0  0  2 45  0  0  0  1  2  0]
 [ 0  0  0  0 39  0  0  0  2  9]
 [ 0  0  0  3  1 39  1  0  4  2]
 [ 0  1  0  0  0  2 47  0  0  0]
 [ 0  0  0  0  0  0  0 44  1  5]
 [ 0  0  1  2  0  2  0  0 45  0]
 [ 0  0  0  0  2  0  0  2  0 46]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.66      0.80        50
           1       0.98      1.00      0.99        50
           2       0.92      0.90      0.91        50
           3       0.88      0.90      0.89        50
           4       0.91      0.78      0.84        50
           5       0.74      0.78      0.76        50
           6       0.92      0.94      0.93        50
           7       0.94      0.88      0.91        50
           8       0.76      0.90      0.83        50
           9       0.73      0.92      0.81        50

    accuracy                           0.87       500
   macro avg       0.88      0.87      0.87       500
weighted avg       0.88      0.87      0.87       500

Generation time: 1.0920 seconds
Coverage=0.840
Avg set size=0.94
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.600    | 0.88     | 0.64    | 2.24     | 0.74    | 0.14    | 0.05    | 0      
1     | 50   | 0.689    | 0.98     | 0.98    | 1.00     | 0.69    | 0.11    | 0.08    | 1      
2     | 50   | 0.804    | 0.94     | 0.88    | 1.24     | 0.82    | 0.09    | 0.03    | 2      
3     | 50   | 0.790    | 0.94     | 0.90    | 1.18     | 0.81    | 0.10    | 0.05    | 3      
4     | 50   | 0.549    | 0.88     | 0.70    | 1.28     | 0.58    | 0.23    | 0.08    | 4      
5     | 50   | 0.535    | 0.86     | 0.76    | 1.48     | 0.59    | 0.22    | 0.08    | 5      
6     | 50   | 0.840    | 0.96     | 0.92    | 1.16     | 0.86    | 0.06    | 0.03    | 6      
7     | 50   | 0.766    | 1.04     | 0.88    | 1.16     | 0.81    | 0.14    | 0.02    | 7      
8     | 50   | 0.628    | 0.94     | 0.84    | 1.12     | 0.64    | 0.19    | 0.08    | 8      
9     | 50   | 0.693    | 0.96     | 0.90    | 1.08     | 0.70    | 0.18    | 0.06    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.544
Confusion matrix (rows=true, cols=pred):
 [[38  0  0  0  0  8  0  0  4  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 46  0  0  0  0  0  4  0]
 [ 0  0  1 27  0  1  0  0 21  0]
 [ 0  0  0  0  1  0  0  0 35 14]
 [ 0  0  0  0  0 14  0  0 36  0]
 [ 0  0  5  0  0  0 31  0 14  0]
 [ 0  0  0  0  0  0  0 20  6 24]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  5 45]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.76      0.86        50
           1       0.00      0.00      0.00        50
           2       0.88      0.92      0.90        50
           3       1.00      0.54      0.70        50
           4       1.00      0.02      0.04        50
           5       0.61      0.28      0.38        50
           6       1.00      0.62      0.77        50
           7       1.00      0.40      0.57        50
           8       0.22      1.00      0.36        50
           9       0.54      0.90      0.68        50

    accuracy                           0.54       500
   macro avg       0.73      0.54      0.53       500
weighted avg       0.73      0.54      0.53       500

Generation time: 0.9827 seconds
Coverage=0.226
Avg set size=0.23
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.406    | 0.48     | 0.46    | 1.50     | 0.44    | 0.22    | 0.14    | 0      
1     | 50   | 0.131    | 0.00     | 0.00    | 2.92     | 0.33    | 0.14    | 0.12    | 8      
2     | 50   | 0.450    | 0.52     | 0.52    | 1.08     | 0.45    | 0.22    | 0.10    | 2      
3     | 50   | 0.298    | 0.08     | 0.06    | 1.52     | 0.34    | 0.24    | 0.17    | 3      
4     | 50   | 0.164    | 0.00     | 0.00    | 2.96     | 0.27    | 0.21    | 0.16    | 8      
5     | 50   | 0.303    | 0.04     | 0.02    | 1.72     | 0.36    | 0.28    | 0.10    | 8      
6     | 50   | 0.279    | 0.10     | 0.08    | 1.54     | 0.32    | 0.20    | 0.16    | 6      
7     | 50   | 0.282    | 0.08     | 0.08    | 1.82     | 0.32    | 0.26    | 0.16    | 9      
8     | 50   | 0.485    | 0.94     | 0.94    | 1.00     | 0.49    | 0.16    | 0.10    | 8      
9     | 50   | 0.344    | 0.10     | 0.10    | 1.10     | 0.35    | 0.23    | 0.12    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.6156 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.068    | 0.00     | 0.00    | 5.52     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.039    | 0.00     | 0.00    | 8.80     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.158    | 0.00     | 0.00    | 2.34     | 0.34    | 0.16    | 0.15    | 8      
3     | 50   | 0.117    | 0.00     | 0.00    | 3.62     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.04     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.185    | 0.00     | 0.00    | 2.00     | 0.35    | 0.19    | 0.10    | 8      
6     | 50   | 0.071    | 0.00     | 0.00    | 5.62     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 5.94     | 0.33    | 0.15    | 0.14    | 8      
8     | 50   | 0.372    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.142    | 0.00     | 0.00    | 2.68     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=10 | std=2.0 ---

== PCA [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.542
Confusion matrix (rows=true, cols=pred):
 [[22  0  6  3  0 11  2  3  2  1]
 [ 0 26  2  2  0  4  0  0 16  0]
 [ 1  1 25  2  2  8  1  3  7  0]
 [ 0  0  9 22  1 11  0  1  3  3]
 [ 0  3  1  0 32  8  0  1  2  3]
 [ 0  0  4  2  3 30  3  0  7  1]
 [ 1  1  5  0  3  7 30  0  3  0]
 [ 0  1  3  0  2  0  0 39  0  5]
 [ 2  1  2  3  4 10  1  0 23  4]
 [ 2  0  0  2  1  7  0 12  4 22]]
Classification report:
               precision    recall  f1-score   support

           0       0.79      0.44      0.56        50
           1       0.79      0.52      0.63        50
           2       0.44      0.50      0.47        50
           3       0.61      0.44      0.51        50
           4       0.67      0.64      0.65        50
           5       0.31      0.60      0.41        50
           6       0.81      0.60      0.69        50
           7       0.66      0.78      0.72        50
           8       0.34      0.46      0.39        50
           9       0.56      0.44      0.49        50

    accuracy                           0.54       500
   macro avg       0.60      0.54      0.55       500
weighted avg       0.60      0.54      0.55       500

Generation time: 0.6297 seconds
Coverage=0.484
Avg set size=0.77
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.354    | 0.74     | 0.36    | 1.82     | 0.53    | 0.23    | 0.11    | 0      
1     | 50   | 0.411    | 0.70     | 0.48    | 1.92     | 0.52    | 0.22    | 0.10    | 1      
2     | 50   | 0.429    | 0.82     | 0.48    | 2.44     | 0.63    | 0.17    | 0.09    | 2      
3     | 50   | 0.393    | 0.72     | 0.38    | 2.54     | 0.60    | 0.18    | 0.09    | 3      
4     | 50   | 0.503    | 0.78     | 0.56    | 1.80     | 0.61    | 0.17    | 0.09    | 4      
5     | 50   | 0.486    | 0.88     | 0.60    | 1.88     | 0.61    | 0.21    | 0.08    | 5      
6     | 50   | 0.490    | 0.80     | 0.56    | 1.66     | 0.60    | 0.20    | 0.09    | 6      
7     | 50   | 0.595    | 0.94     | 0.74    | 1.24     | 0.64    | 0.19    | 0.08    | 7      
8     | 50   | 0.361    | 0.68     | 0.36    | 2.20     | 0.54    | 0.22    | 0.10    | 8      
9     | 50   | 0.360    | 0.66     | 0.32    | 1.96     | 0.52    | 0.22    | 0.11    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.752
Confusion matrix (rows=true, cols=pred):
 [[44  0  0  0  0  3  0  0  3  0]
 [ 0 44  0  0  0  0  0  0  6  0]
 [ 1  0 47  0  0  0  0  0  2  0]
 [ 1  0  7 29  0  6  0  0  7  0]
 [ 0  0  0  0 24  0  0  0  7 19]
 [ 1  0  0  0  0 32  0  0 17  0]
 [ 0  0 11  0  0  2 34  0  3  0]
 [ 1  0  1  0  0  3  0 28  7 10]
 [ 0  0  1  0  0  0  0  0 49  0]
 [ 0  0  0  0  0  1  0  0  4 45]]
Classification report:
               precision    recall  f1-score   support

           0       0.92      0.88      0.90        50
           1       1.00      0.88      0.94        50
           2       0.70      0.94      0.80        50
           3       1.00      0.58      0.73        50
           4       1.00      0.48      0.65        50
           5       0.68      0.64      0.66        50
           6       1.00      0.68      0.81        50
           7       1.00      0.56      0.72        50
           8       0.47      0.98      0.63        50
           9       0.61      0.90      0.73        50

    accuracy                           0.75       500
   macro avg       0.84      0.75      0.76       500
weighted avg       0.84      0.75      0.76       500

Generation time: 0.9770 seconds
Coverage=0.484
Avg set size=0.52
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.528    | 0.66     | 0.66    | 1.28     | 0.54    | 0.19    | 0.11    | 0      
1     | 50   | 0.552    | 0.76     | 0.76    | 1.12     | 0.56    | 0.17    | 0.10    | 1      
2     | 50   | 0.478    | 0.64     | 0.64    | 1.06     | 0.48    | 0.20    | 0.10    | 2      
3     | 50   | 0.339    | 0.38     | 0.34    | 2.00     | 0.40    | 0.23    | 0.16    | 3      
4     | 50   | 0.307    | 0.24     | 0.18    | 1.78     | 0.36    | 0.24    | 0.15    | 4      
5     | 50   | 0.357    | 0.30     | 0.20    | 1.36     | 0.39    | 0.27    | 0.12    | 5      
6     | 50   | 0.366    | 0.50     | 0.44    | 1.70     | 0.42    | 0.18    | 0.13    | 6      
7     | 50   | 0.299    | 0.24     | 0.18    | 2.06     | 0.34    | 0.26    | 0.12    | 7      
8     | 50   | 0.524    | 0.78     | 0.78    | 1.02     | 0.53    | 0.17    | 0.10    | 8      
9     | 50   | 0.504    | 0.66     | 0.66    | 1.22     | 0.52    | 0.16    | 0.11    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.766
Confusion matrix (rows=true, cols=pred):
 [[44  0  2  0  0  4  0  0  0  0]
 [ 0 48  0  0  0  0  0  0  2  0]
 [ 0  0 46  0  0  0  0  0  4  0]
 [ 1  0  8 26  0  4  0  0 11  0]
 [ 2  0  1  0 29  0  0  0  5 13]
 [ 0  0  2  1  0 34  0  0 13  0]
 [ 0  0 11  0  0  1 35  0  3  0]
 [ 0  0  2  0  0  5  0 29  6  8]
 [ 0  0  0  0  0  2  0  0 47  1]
 [ 0  0  1  0  0  1  0  0  3 45]]
Classification report:
               precision    recall  f1-score   support

           0       0.94      0.88      0.91        50
           1       1.00      0.96      0.98        50
           2       0.63      0.92      0.75        50
           3       0.96      0.52      0.68        50
           4       1.00      0.58      0.73        50
           5       0.67      0.68      0.67        50
           6       1.00      0.70      0.82        50
           7       1.00      0.58      0.73        50
           8       0.50      0.94      0.65        50
           9       0.67      0.90      0.77        50

    accuracy                           0.77       500
   macro avg       0.84      0.77      0.77       500
weighted avg       0.84      0.77      0.77       500

Generation time: 0.9901 seconds
Coverage=0.518
Avg set size=0.54
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.510    | 0.70     | 0.70    | 1.16     | 0.52    | 0.20    | 0.11    | 0      
1     | 50   | 0.695    | 0.84     | 0.84    | 1.04     | 0.70    | 0.12    | 0.07    | 1      
2     | 50   | 0.442    | 0.58     | 0.58    | 1.08     | 0.45    | 0.21    | 0.10    | 2      
3     | 50   | 0.319    | 0.38     | 0.30    | 2.18     | 0.40    | 0.23    | 0.16    | 3      
4     | 50   | 0.350    | 0.32     | 0.28    | 1.70     | 0.39    | 0.25    | 0.13    | 4      
5     | 50   | 0.365    | 0.34     | 0.30    | 1.38     | 0.39    | 0.27    | 0.11    | 5      
6     | 50   | 0.413    | 0.52     | 0.52    | 1.68     | 0.45    | 0.18    | 0.12    | 6      
7     | 50   | 0.297    | 0.24     | 0.18    | 2.20     | 0.34    | 0.26    | 0.12    | 7      
8     | 50   | 0.552    | 0.84     | 0.84    | 1.06     | 0.56    | 0.16    | 0.09    | 8      
9     | 50   | 0.514    | 0.66     | 0.64    | 1.42     | 0.54    | 0.17    | 0.11    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.682
Confusion matrix (rows=true, cols=pred):
 [[42  0  0  0  0  4  0  0  4  0]
 [ 0 48  0  0  0  0  0  0  2  0]
 [ 0  0 34  0  0  3  0  0 13  0]
 [ 0  0  1 22  0  5  0  0 22  0]
 [ 1  0  0  0 28  0  0  0 11 10]
 [ 0  0  3  0  0 29  0  0 18  0]
 [ 1  0  6  0  0  3 34  0  6  0]
 [ 0  0  2  0  0  7  0 17 12 12]
 [ 0  0  1  0  0  0  0  0 48  1]
 [ 0  0  3  0  0  2  0  0  6 39]]
Classification report:
               precision    recall  f1-score   support

           0       0.95      0.84      0.89        50
           1       1.00      0.96      0.98        50
           2       0.68      0.68      0.68        50
           3       1.00      0.44      0.61        50
           4       1.00      0.56      0.72        50
           5       0.55      0.58      0.56        50
           6       1.00      0.68      0.81        50
           7       1.00      0.34      0.51        50
           8       0.34      0.96      0.50        50
           9       0.63      0.78      0.70        50

    accuracy                           0.68       500
   macro avg       0.81      0.68      0.70       500
weighted avg       0.81      0.68      0.70       500

Generation time: 1.0008 seconds
Coverage=0.374
Avg set size=0.39
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.389    | 0.36     | 0.36    | 1.36     | 0.40    | 0.22    | 0.15    | 0      
1     | 50   | 0.739    | 0.88     | 0.86    | 1.04     | 0.75    | 0.09    | 0.06    | 1      
2     | 50   | 0.316    | 0.14     | 0.14    | 1.40     | 0.34    | 0.24    | 0.13    | 2      
3     | 50   | 0.266    | 0.22     | 0.16    | 2.22     | 0.35    | 0.23    | 0.17    | 3      
4     | 50   | 0.327    | 0.32     | 0.30    | 2.12     | 0.38    | 0.25    | 0.14    | 4      
5     | 50   | 0.328    | 0.10     | 0.10    | 1.48     | 0.35    | 0.28    | 0.11    | 5      
6     | 50   | 0.327    | 0.32     | 0.32    | 1.92     | 0.37    | 0.19    | 0.15    | 6      
7     | 50   | 0.224    | 0.12     | 0.08    | 3.26     | 0.32    | 0.25    | 0.13    | 7      
8     | 50   | 0.540    | 0.80     | 0.80    | 1.04     | 0.54    | 0.15    | 0.10    | 8      
9     | 50   | 0.466    | 0.62     | 0.62    | 1.66     | 0.50    | 0.17    | 0.11    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.4195 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.089    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.050    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.043    | 0.00     | 0.00    | 7.44     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.910
Confusion matrix (rows=true, cols=pred):
 [[39  0  2  0  0  4  1  0  4  0]
 [ 0 48  0  0  0  0  0  0  2  0]
 [ 0  0 46  2  0  1  0  0  1  0]
 [ 0  0  1 48  0  0  0  1  0  0]
 [ 0  0  1  0 44  0  0  0  0  5]
 [ 0  1  0  2  2 41  0  0  4  0]
 [ 0  0  0  1  0  0 48  0  1  0]
 [ 0  0  0  0  0  0  0 49  0  1]
 [ 0  1  2  3  0  0  0  0 43  1]
 [ 0  0  0  0  0  0  0  0  1 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.78      0.88        50
           1       0.96      0.96      0.96        50
           2       0.88      0.92      0.90        50
           3       0.86      0.96      0.91        50
           4       0.96      0.88      0.92        50
           5       0.89      0.82      0.85        50
           6       0.98      0.96      0.97        50
           7       0.98      0.98      0.98        50
           8       0.77      0.86      0.81        50
           9       0.88      0.98      0.92        50

    accuracy                           0.91       500
   macro avg       0.92      0.91      0.91       500
weighted avg       0.92      0.91      0.91       500

Generation time: 1.2012 seconds
Coverage=0.820
Avg set size=0.86
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.682    | 0.74     | 0.68    | 1.66     | 0.73    | 0.13    | 0.05    | 0      
1     | 50   | 0.550    | 0.76     | 0.76    | 1.06     | 0.55    | 0.18    | 0.10    | 1      
2     | 50   | 0.810    | 0.94     | 0.90    | 1.14     | 0.82    | 0.08    | 0.04    | 2      
3     | 50   | 0.807    | 0.92     | 0.92    | 1.04     | 0.81    | 0.09    | 0.04    | 3      
4     | 50   | 0.544    | 0.78     | 0.70    | 1.14     | 0.57    | 0.20    | 0.08    | 4      
5     | 50   | 0.579    | 0.88     | 0.74    | 1.40     | 0.62    | 0.20    | 0.09    | 5      
6     | 50   | 0.777    | 0.90     | 0.90    | 1.14     | 0.79    | 0.11    | 0.05    | 6      
7     | 50   | 0.816    | 0.94     | 0.92    | 1.02     | 0.82    | 0.12    | 0.03    | 7      
8     | 50   | 0.602    | 0.84     | 0.78    | 1.16     | 0.62    | 0.18    | 0.08    | 8      
9     | 50   | 0.671    | 0.90     | 0.90    | 1.02     | 0.68    | 0.15    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=3.60332e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0862 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 7.96     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=7.83564e-09): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.786
Confusion matrix (rows=true, cols=pred):
 [[29  0  4  1  1  8  2  1  4  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 1  0 41  0  1  1  3  1  2  0]
 [ 0  0  4 42  0  0  0  1  2  1]
 [ 0  0  0  0 33  1  2  0  3 11]
 [ 0  1  1  3  2 36  1  2  2  2]
 [ 0  1  0  0  0  4 44  0  1  0]
 [ 0  0  0  0  0  0  0 40  2  8]
 [ 0  3  2  4  0  4  0  0 37  0]
 [ 0  1  0  0  4  0  0  4  0 41]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.58      0.72        50
           1       0.89      1.00      0.94        50
           2       0.79      0.82      0.80        50
           3       0.84      0.84      0.84        50
           4       0.80      0.66      0.73        50
           5       0.67      0.72      0.69        50
           6       0.85      0.88      0.86        50
           7       0.82      0.80      0.81        50
           8       0.70      0.74      0.72        50
           9       0.65      0.82      0.73        50

    accuracy                           0.79       500
   macro avg       0.80      0.79      0.78       500
weighted avg       0.80      0.79      0.78       500

Generation time: 1.6423 seconds
Coverage=0.754
Avg set size=0.90
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.546    | 0.92     | 0.58    | 2.82     | 0.76    | 0.12    | 0.05    | 0      
1     | 50   | 0.666    | 0.90     | 0.90    | 1.00     | 0.67    | 0.12    | 0.08    | 1      
2     | 50   | 0.741    | 0.94     | 0.82    | 1.54     | 0.80    | 0.10    | 0.04    | 2      
3     | 50   | 0.759    | 0.90     | 0.82    | 1.38     | 0.81    | 0.11    | 0.05    | 3      
4     | 50   | 0.494    | 0.78     | 0.56    | 1.54     | 0.57    | 0.21    | 0.09    | 4      
5     | 50   | 0.522    | 0.84     | 0.68    | 1.74     | 0.62    | 0.20    | 0.09    | 5      
6     | 50   | 0.771    | 0.96     | 0.86    | 1.34     | 0.81    | 0.09    | 0.04    | 6      
7     | 50   | 0.741    | 0.94     | 0.80    | 1.42     | 0.81    | 0.13    | 0.03    | 7      
8     | 50   | 0.560    | 0.86     | 0.70    | 1.32     | 0.62    | 0.21    | 0.08    | 8      
9     | 50   | 0.659    | 0.98     | 0.82    | 1.18     | 0.70    | 0.18    | 0.06    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.516
Confusion matrix (rows=true, cols=pred):
 [[36  0  0  0  0  9  0  0  5  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 41  0  0  0  0  0  9  0]
 [ 0  0  1 26  0  2  0  0 21  0]
 [ 0  0  0  0  2  0  0  0 31 17]
 [ 0  0  0  0  0 17  0  0 33  0]
 [ 0  0  6  0  0  1 28  0 15  0]
 [ 0  0  0  0  0  0  0 19 10 21]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 11 39]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.72      0.84        50
           1       0.00      0.00      0.00        50
           2       0.85      0.82      0.84        50
           3       1.00      0.52      0.68        50
           4       1.00      0.04      0.08        50
           5       0.59      0.34      0.43        50
           6       1.00      0.56      0.72        50
           7       1.00      0.38      0.55        50
           8       0.21      1.00      0.35        50
           9       0.51      0.78      0.61        50

    accuracy                           0.52       500
   macro avg       0.72      0.52      0.51       500
weighted avg       0.72      0.52      0.51       500

Generation time: 0.9858 seconds
Coverage=0.232
Avg set size=0.24
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.409    | 0.48     | 0.46    | 1.84     | 0.46    | 0.21    | 0.13    | 0      
1     | 50   | 0.131    | 0.00     | 0.00    | 3.20     | 0.33    | 0.15    | 0.13    | 8      
2     | 50   | 0.437    | 0.52     | 0.52    | 1.22     | 0.45    | 0.21    | 0.11    | 2      
3     | 50   | 0.302    | 0.22     | 0.20    | 1.72     | 0.36    | 0.23    | 0.17    | 3      
4     | 50   | 0.162    | 0.00     | 0.00    | 2.94     | 0.27    | 0.21    | 0.16    | 8      
5     | 50   | 0.297    | 0.08     | 0.02    | 1.70     | 0.36    | 0.27    | 0.11    | 8      
6     | 50   | 0.275    | 0.12     | 0.10    | 1.86     | 0.34    | 0.19    | 0.15    | 6      
7     | 50   | 0.283    | 0.10     | 0.10    | 1.98     | 0.33    | 0.25    | 0.15    | 9      
8     | 50   | 0.468    | 0.76     | 0.76    | 1.00     | 0.47    | 0.17    | 0.10    | 8      
9     | 50   | 0.344    | 0.16     | 0.16    | 1.22     | 0.35    | 0.22    | 0.12    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=10² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0004 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.069    | 0.00     | 0.00    | 5.54     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.038    | 0.00     | 0.00    | 8.76     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.157    | 0.00     | 0.00    | 2.42     | 0.34    | 0.16    | 0.14    | 8      
3     | 50   | 0.117    | 0.00     | 0.00    | 3.64     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.02     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.185    | 0.00     | 0.00    | 2.00     | 0.35    | 0.19    | 0.11    | 8      
6     | 50   | 0.071    | 0.00     | 0.00    | 5.56     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 5.92     | 0.33    | 0.15    | 0.14    | 8      
8     | 50   | 0.371    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.142    | 0.00     | 0.00    | 2.58     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=12 | std=1.0 ---

== PCA [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.822
Confusion matrix (rows=true, cols=pred):
 [[36  0  3  0  0 11  0  0  0  0]
 [ 0 46  0  0  0  0  0  0  4  0]
 [ 0  0 32  2  0  9  1  0  3  3]
 [ 0  0  5 32  0  6  0  1  6  0]
 [ 0  0  0  0 43  2  0  1  2  2]
 [ 0  0  0  1  2 45  0  0  2  0]
 [ 0  0  0  0  1  0 47  0  2  0]
 [ 0  0  0  0  1  0  0 42  0  7]
 [ 0  0  2  0  0  1  1  0 44  2]
 [ 0  0  0  0  0  1  0  0  5 44]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.72      0.84        50
           1       1.00      0.92      0.96        50
           2       0.76      0.64      0.70        50
           3       0.91      0.64      0.75        50
           4       0.91      0.86      0.89        50
           5       0.60      0.90      0.72        50
           6       0.96      0.94      0.95        50
           7       0.95      0.84      0.89        50
           8       0.65      0.88      0.75        50
           9       0.76      0.88      0.81        50

    accuracy                           0.82       500
   macro avg       0.85      0.82      0.83       500
weighted avg       0.85      0.82      0.83       500

Generation time: 0.6755 seconds
Coverage=0.690
Avg set size=0.75
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.459    | 0.68     | 0.56    | 1.36     | 0.52    | 0.22    | 0.11    | 0      
1     | 50   | 0.516    | 0.78     | 0.76    | 1.10     | 0.53    | 0.22    | 0.10    | 1      
2     | 50   | 0.435    | 0.60     | 0.56    | 2.08     | 0.52    | 0.17    | 0.10    | 2      
3     | 50   | 0.416    | 0.62     | 0.50    | 1.60     | 0.47    | 0.21    | 0.11    | 3      
4     | 50   | 0.597    | 0.74     | 0.70    | 1.16     | 0.61    | 0.16    | 0.10    | 4      
5     | 50   | 0.524    | 0.74     | 0.70    | 1.12     | 0.54    | 0.20    | 0.09    | 5      
6     | 50   | 0.546    | 0.74     | 0.72    | 1.08     | 0.56    | 0.16    | 0.10    | 6      
7     | 50   | 0.578    | 1.00     | 0.82    | 1.16     | 0.60    | 0.24    | 0.06    | 7      
8     | 50   | 0.537    | 0.76     | 0.76    | 1.14     | 0.55    | 0.20    | 0.09    | 8      
9     | 50   | 0.580    | 0.84     | 0.82    | 1.14     | 0.59    | 0.17    | 0.09    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.848
Confusion matrix (rows=true, cols=pred):
 [[49  0  0  0  0  1  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  3 33  0  1  0  0 13  0]
 [ 0  0  0  0 29  0  0  0  6 15]
 [ 0  0  0  0  0 40  0  0 10  0]
 [ 0  0  3  0  0  0 44  0  3  0]
 [ 0  0  0  0  0  2  0 31  9  8]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  1 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.98      0.99        50
           1       1.00      1.00      1.00        50
           2       0.89      0.98      0.93        50
           3       1.00      0.66      0.80        50
           4       1.00      0.58      0.73        50
           5       0.91      0.80      0.85        50
           6       1.00      0.88      0.94        50
           7       1.00      0.62      0.77        50
           8       0.54      1.00      0.70        50
           9       0.68      0.98      0.80        50

    accuracy                           0.85       500
   macro avg       0.90      0.85      0.85       500
weighted avg       0.90      0.85      0.85       500

Generation time: 1.1315 seconds
Coverage=0.512
Avg set size=0.51
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.523    | 0.70     | 0.70    | 1.02     | 0.52    | 0.20    | 0.12    | 0      
1     | 50   | 0.564    | 0.96     | 0.96    | 1.00     | 0.56    | 0.19    | 0.08    | 1      
2     | 50   | 0.497    | 0.82     | 0.82    | 1.02     | 0.50    | 0.19    | 0.09    | 2      
3     | 50   | 0.329    | 0.16     | 0.16    | 1.44     | 0.35    | 0.25    | 0.18    | 3      
4     | 50   | 0.288    | 0.06     | 0.06    | 1.68     | 0.31    | 0.26    | 0.17    | 4      
5     | 50   | 0.373    | 0.20     | 0.20    | 1.20     | 0.38    | 0.30    | 0.09    | 5      
6     | 50   | 0.392    | 0.32     | 0.32    | 1.16     | 0.40    | 0.19    | 0.14    | 6      
7     | 50   | 0.290    | 0.12     | 0.10    | 1.78     | 0.31    | 0.26    | 0.13    | 7      
8     | 50   | 0.625    | 1.00     | 1.00    | 1.00     | 0.62    | 0.12    | 0.08    | 8      
9     | 50   | 0.490    | 0.80     | 0.80    | 1.02     | 0.49    | 0.16    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.858
Confusion matrix (rows=true, cols=pred):
 [[50  0  0  0  0  0  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  2 31  0  4  0  0 13  0]
 [ 0  0  1  0 35  0  0  0  5  9]
 [ 0  0  0  0  0 42  0  0  8  0]
 [ 0  0  4  0  0  1 43  0  2  0]
 [ 0  0  0  0  0  1  0 31 10  8]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  2 48]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00        50
           1       1.00      1.00      1.00        50
           2       0.88      0.98      0.92        50
           3       1.00      0.62      0.77        50
           4       1.00      0.70      0.82        50
           5       0.88      0.84      0.86        50
           6       1.00      0.86      0.92        50
           7       1.00      0.62      0.77        50
           8       0.55      1.00      0.71        50
           9       0.74      0.96      0.83        50

    accuracy                           0.86       500
   macro avg       0.90      0.86      0.86       500
weighted avg       0.90      0.86      0.86       500

Generation time: 1.3704 seconds
Coverage=0.540
Avg set size=0.54
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.516    | 0.72     | 0.72    | 1.00     | 0.52    | 0.20    | 0.11    | 0      
1     | 50   | 0.714    | 1.00     | 1.00    | 1.00     | 0.71    | 0.12    | 0.06    | 1      
2     | 50   | 0.453    | 0.60     | 0.60    | 1.02     | 0.45    | 0.21    | 0.11    | 2      
3     | 50   | 0.335    | 0.18     | 0.18    | 1.54     | 0.36    | 0.25    | 0.18    | 3      
4     | 50   | 0.322    | 0.24     | 0.24    | 1.50     | 0.35    | 0.26    | 0.15    | 4      
5     | 50   | 0.383    | 0.22     | 0.22    | 1.16     | 0.39    | 0.29    | 0.09    | 5      
6     | 50   | 0.422    | 0.48     | 0.48    | 1.24     | 0.43    | 0.17    | 0.14    | 6      
7     | 50   | 0.293    | 0.16     | 0.16    | 1.90     | 0.32    | 0.26    | 0.13    | 7      
8     | 50   | 0.663    | 0.98     | 0.98    | 1.00     | 0.66    | 0.11    | 0.07    | 8      
9     | 50   | 0.518    | 0.82     | 0.82    | 1.06     | 0.52    | 0.16    | 0.12    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.704
Confusion matrix (rows=true, cols=pred):
 [[41  0  0  0  0  9  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 34  0  0  0  0  0 16  0]
 [ 0  0  2 20  0  2  0  0 26  0]
 [ 0  0  1  0 24  0  0  0 18  7]
 [ 0  0  0  0  0 37  0  0 13  0]
 [ 0  0  4  0  0  2 35  0  9  0]
 [ 0  0  0  0  0  2  0 15 21 12]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  4 46]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.82      0.90        50
           1       1.00      1.00      1.00        50
           2       0.83      0.68      0.75        50
           3       1.00      0.40      0.57        50
           4       1.00      0.48      0.65        50
           5       0.71      0.74      0.73        50
           6       1.00      0.70      0.82        50
           7       1.00      0.30      0.46        50
           8       0.32      1.00      0.48        50
           9       0.71      0.92      0.80        50

    accuracy                           0.70       500
   macro avg       0.86      0.70      0.72       500
weighted avg       0.86      0.70      0.72       500

Generation time: 1.0694 seconds
Coverage=0.328
Avg set size=0.33
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.369    | 0.26     | 0.26    | 1.30     | 0.38    | 0.23    | 0.16    | 0      
1     | 50   | 0.775    | 1.00     | 1.00    | 1.00     | 0.77    | 0.10    | 0.05    | 1      
2     | 50   | 0.304    | 0.02     | 0.02    | 1.34     | 0.32    | 0.25    | 0.14    | 2      
3     | 50   | 0.263    | 0.04     | 0.04    | 2.10     | 0.32    | 0.25    | 0.19    | 8      
4     | 50   | 0.282    | 0.14     | 0.14    | 1.94     | 0.33    | 0.24    | 0.16    | 4      
5     | 50   | 0.340    | 0.04     | 0.04    | 1.26     | 0.35    | 0.29    | 0.10    | 5      
6     | 50   | 0.303    | 0.16     | 0.16    | 1.70     | 0.33    | 0.19    | 0.16    | 6      
7     | 50   | 0.209    | 0.06     | 0.06    | 3.08     | 0.28    | 0.23    | 0.14    | 8      
8     | 50   | 0.627    | 0.98     | 0.98    | 1.00     | 0.63    | 0.12    | 0.08    | 8      
9     | 50   | 0.465    | 0.58     | 0.58    | 1.08     | 0.47    | 0.17    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9239 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 7.98     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.988
Confusion matrix (rows=true, cols=pred):
 [[49  0  0  0  0  1  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 50  0  0  0  0  0  0]
 [ 0  0  0  0 47  0  0  0  0  3]
 [ 0  0  0  0  0 49  0  0  1  0]
 [ 0  0  0  0  0  0 50  0  0  0]
 [ 0  0  0  0  0  0  0 49  0  1]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.98      0.99        50
           1       1.00      1.00      1.00        50
           2       1.00      1.00      1.00        50
           3       1.00      1.00      1.00        50
           4       1.00      0.94      0.97        50
           5       0.98      0.98      0.98        50
           6       1.00      1.00      1.00        50
           7       1.00      0.98      0.99        50
           8       0.98      1.00      0.99        50
           9       0.93      1.00      0.96        50

    accuracy                           0.99       500
   macro avg       0.99      0.99      0.99       500
weighted avg       0.99      0.99      0.99       500

Generation time: 1.4367 seconds
Coverage=0.974
Avg set size=0.98
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.841    | 0.94     | 0.94    | 1.06     | 0.84    | 0.08    | 0.04    | 0      
1     | 50   | 0.599    | 1.00     | 1.00    | 1.00     | 0.60    | 0.16    | 0.08    | 1      
2     | 50   | 0.921    | 1.00     | 1.00    | 1.00     | 0.92    | 0.03    | 0.02    | 2      
3     | 50   | 0.804    | 0.98     | 0.98    | 1.00     | 0.80    | 0.09    | 0.05    | 3      
4     | 50   | 0.609    | 0.94     | 0.90    | 1.06     | 0.61    | 0.20    | 0.07    | 4      
5     | 50   | 0.639    | 0.96     | 0.96    | 1.02     | 0.64    | 0.21    | 0.06    | 5      
6     | 50   | 0.867    | 0.98     | 0.98    | 1.00     | 0.87    | 0.06    | 0.03    | 6      
7     | 50   | 0.821    | 1.04     | 1.00    | 1.02     | 0.82    | 0.13    | 0.02    | 7      
8     | 50   | 0.740    | 0.98     | 0.98    | 1.00     | 0.74    | 0.11    | 0.06    | 8      
9     | 50   | 0.684    | 1.00     | 1.00    | 1.00     | 0.68    | 0.14    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=4.1806e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.1755 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=1.095e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.982
Confusion matrix (rows=true, cols=pred):
 [[49  0  0  0  0  1  0  0  0  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 50  0  0  0  0  0  0  0]
 [ 0  0  0 50  0  0  0  0  0  0]
 [ 0  0  0  0 48  0  0  0  0  2]
 [ 0  0  0  2  0 47  0  0  1  0]
 [ 0  0  0  0  0  0 50  0  0  0]
 [ 0  0  0  0  0  0  0 48  0  2]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  1  0 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.98      0.99        50
           1       1.00      1.00      1.00        50
           2       1.00      1.00      1.00        50
           3       0.96      1.00      0.98        50
           4       1.00      0.96      0.98        50
           5       0.98      0.94      0.96        50
           6       1.00      1.00      1.00        50
           7       0.98      0.96      0.97        50
           8       0.98      1.00      0.99        50
           9       0.92      0.98      0.95        50

    accuracy                           0.98       500
   macro avg       0.98      0.98      0.98       500
weighted avg       0.98      0.98      0.98       500

Generation time: 1.1339 seconds
Coverage=0.972
Avg set size=0.99
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.861    | 1.00     | 0.98    | 1.02     | 0.86    | 0.09    | 0.02    | 0      
1     | 50   | 0.808    | 1.00     | 1.00    | 1.00     | 0.81    | 0.08    | 0.04    | 1      
2     | 50   | 0.949    | 1.00     | 1.00    | 1.00     | 0.95    | 0.03    | 0.01    | 2      
3     | 50   | 0.927    | 0.98     | 0.98    | 1.00     | 0.93    | 0.04    | 0.02    | 3      
4     | 50   | 0.761    | 1.00     | 0.96    | 1.04     | 0.77    | 0.15    | 0.03    | 4      
5     | 50   | 0.740    | 0.96     | 0.90    | 1.06     | 0.75    | 0.15    | 0.05    | 5      
6     | 50   | 0.955    | 1.00     | 1.00    | 1.00     | 0.95    | 0.02    | 0.01    | 6      
7     | 50   | 0.869    | 1.00     | 0.96    | 1.04     | 0.87    | 0.10    | 0.01    | 7      
8     | 50   | 0.801    | 0.96     | 0.96    | 1.00     | 0.80    | 0.09    | 0.04    | 8      
9     | 50   | 0.780    | 0.98     | 0.98    | 1.02     | 0.78    | 0.12    | 0.05    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.550
Confusion matrix (rows=true, cols=pred):
 [[44  0  0  0  0  3  0  0  3  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0 48  0  0  0  0  0  2  0]
 [ 0  0  0 27  0  0  0  0 23  0]
 [ 0  0  0  0  0  0  0  0 35 15]
 [ 0  0  0  0  0 10  0  0 40  0]
 [ 0  0  3  0  0  0 37  0 10  0]
 [ 0  0  0  0  0  0  0 17  3 30]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  8 42]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.88      0.94        50
           1       0.00      0.00      0.00        50
           2       0.94      0.96      0.95        50
           3       1.00      0.54      0.70        50
           4       0.00      0.00      0.00        50
           5       0.77      0.20      0.32        50
           6       1.00      0.74      0.85        50
           7       1.00      0.34      0.51        50
           8       0.22      1.00      0.36        50
           9       0.48      0.84      0.61        50

    accuracy                           0.55       500
   macro avg       0.64      0.55      0.52       500
weighted avg       0.64      0.55      0.52       500

Generation time: 1.0416 seconds
Coverage=0.224
Avg set size=0.22
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.422    | 0.42     | 0.42    | 1.16     | 0.43    | 0.22    | 0.15    | 0      
1     | 50   | 0.132    | 0.00     | 0.00    | 2.34     | 0.33    | 0.14    | 0.12    | 8      
2     | 50   | 0.475    | 0.72     | 0.72    | 1.04     | 0.48    | 0.22    | 0.09    | 2      
3     | 50   | 0.295    | 0.04     | 0.04    | 1.50     | 0.32    | 0.26    | 0.18    | 3      
4     | 50   | 0.164    | 0.00     | 0.00    | 2.96     | 0.26    | 0.22    | 0.16    | 8      
5     | 50   | 0.299    | 0.00     | 0.00    | 1.80     | 0.35    | 0.29    | 0.09    | 8      
6     | 50   | 0.287    | 0.08     | 0.08    | 1.38     | 0.30    | 0.21    | 0.17    | 6      
7     | 50   | 0.277    | 0.02     | 0.02    | 1.86     | 0.31    | 0.26    | 0.16    | 9      
8     | 50   | 0.498    | 0.92     | 0.92    | 1.00     | 0.50    | 0.15    | 0.10    | 8      
9     | 50   | 0.333    | 0.04     | 0.04    | 1.16     | 0.34    | 0.24    | 0.12    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0261 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.069    | 0.00     | 0.00    | 5.42     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.039    | 0.00     | 0.00    | 8.80     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.162    | 0.00     | 0.00    | 2.26     | 0.34    | 0.16    | 0.15    | 8      
3     | 50   | 0.116    | 0.00     | 0.00    | 3.56     | 0.35    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.02     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.184    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.10    | 8      
6     | 50   | 0.071    | 0.00     | 0.00    | 5.72     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 6.00     | 0.32    | 0.15    | 0.14    | 8      
8     | 50   | 0.374    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.141    | 0.00     | 0.00    | 2.78     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=12 | std=1.5 ---

== PCA [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.692
Confusion matrix (rows=true, cols=pred):
 [[28  0  5  1  0 13  1  1  0  1]
 [ 0 39  0  3  0  1  0  0  7  0]
 [ 2  0 29  2  0 10  2  1  3  1]
 [ 1  0  9 27  0  6  0  1  5  1]
 [ 0  0  0  0 37  4  0  2  4  3]
 [ 1  0  2  5  3 34  0  0  5  0]
 [ 0  0  3  0  2  4 40  0  1  0]
 [ 0  0  1  0  1  0  0 40  0  8]
 [ 0  0  6  1  1  4  2  0 33  3]
 [ 0  0  1  0  0  2  0  3  5 39]]
Classification report:
               precision    recall  f1-score   support

           0       0.88      0.56      0.68        50
           1       1.00      0.78      0.88        50
           2       0.52      0.58      0.55        50
           3       0.69      0.54      0.61        50
           4       0.84      0.74      0.79        50
           5       0.44      0.68      0.53        50
           6       0.89      0.80      0.84        50
           7       0.83      0.80      0.82        50
           8       0.52      0.66      0.58        50
           9       0.70      0.78      0.74        50

    accuracy                           0.69       500
   macro avg       0.73      0.69      0.70       500
weighted avg       0.73      0.69      0.70       500

Generation time: 0.6851 seconds
Coverage=0.592
Avg set size=0.76
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.408    | 0.70     | 0.50    | 1.80     | 0.54    | 0.22    | 0.10    | 0      
1     | 50   | 0.520    | 0.70     | 0.62    | 1.34     | 0.56    | 0.20    | 0.10    | 1      
2     | 50   | 0.449    | 0.72     | 0.52    | 2.48     | 0.59    | 0.17    | 0.09    | 2      
3     | 50   | 0.430    | 0.74     | 0.48    | 1.94     | 0.54    | 0.20    | 0.10    | 3      
4     | 50   | 0.531    | 0.76     | 0.60    | 1.38     | 0.62    | 0.16    | 0.09    | 4      
5     | 50   | 0.496    | 0.72     | 0.56    | 1.50     | 0.56    | 0.20    | 0.09    | 5      
6     | 50   | 0.548    | 0.70     | 0.66    | 1.32     | 0.58    | 0.18    | 0.09    | 6      
7     | 50   | 0.584    | 0.94     | 0.76    | 1.22     | 0.63    | 0.21    | 0.07    | 7      
8     | 50   | 0.449    | 0.78     | 0.54    | 1.48     | 0.53    | 0.24    | 0.10    | 8      
9     | 50   | 0.535    | 0.80     | 0.68    | 1.30     | 0.59    | 0.19    | 0.08    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.784
Confusion matrix (rows=true, cols=pred):
 [[46  0  0  0  0  4  0  0  0  0]
 [ 0 48  0  0  0  0  0  0  2  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  5 30  0  4  0  0 11  0]
 [ 0  0  1  0 26  0  0  0 10 13]
 [ 0  0  1  0  0 36  0  0 13  0]
 [ 0  0  8  0  0  1 34  0  7  0]
 [ 0  0  3  0  0  5  0 27  9  6]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  4 46]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.92      0.96        50
           1       1.00      0.96      0.98        50
           2       0.73      0.98      0.84        50
           3       1.00      0.60      0.75        50
           4       1.00      0.52      0.68        50
           5       0.72      0.72      0.72        50
           6       1.00      0.68      0.81        50
           7       1.00      0.54      0.70        50
           8       0.47      1.00      0.64        50
           9       0.71      0.92      0.80        50

    accuracy                           0.78       500
   macro avg       0.86      0.78      0.79       500
weighted avg       0.86      0.78      0.79       500

Generation time: 1.6887 seconds
Coverage=0.486
Avg set size=0.50
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.507    | 0.60     | 0.60    | 1.12     | 0.51    | 0.20    | 0.13    | 0      
1     | 50   | 0.537    | 0.78     | 0.74    | 1.04     | 0.54    | 0.20    | 0.09    | 1      
2     | 50   | 0.491    | 0.72     | 0.72    | 1.02     | 0.49    | 0.19    | 0.10    | 2      
3     | 50   | 0.325    | 0.26     | 0.20    | 1.68     | 0.38    | 0.24    | 0.17    | 3      
4     | 50   | 0.276    | 0.12     | 0.12    | 1.94     | 0.32    | 0.25    | 0.15    | 4      
5     | 50   | 0.361    | 0.32     | 0.28    | 1.32     | 0.38    | 0.27    | 0.11    | 5      
6     | 50   | 0.386    | 0.34     | 0.34    | 1.56     | 0.41    | 0.19    | 0.13    | 6      
7     | 50   | 0.286    | 0.22     | 0.18    | 2.52     | 0.34    | 0.26    | 0.12    | 7      
8     | 50   | 0.610    | 0.96     | 0.96    | 1.00     | 0.61    | 0.13    | 0.09    | 8      
9     | 50   | 0.480    | 0.72     | 0.72    | 1.18     | 0.49    | 0.17    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.782
Confusion matrix (rows=true, cols=pred):
 [[44  0  0  0  0  5  0  0  1  0]
 [ 0 47  0  0  0  0  0  0  3  0]
 [ 0  0 47  0  0  0  0  0  3  0]
 [ 0  0  4 30  0  4  0  0 12  0]
 [ 1  0  1  0 31  0  0  0  8  9]
 [ 0  0  3  0  0 37  0  0 10  0]
 [ 2  0  4  0  0  2 37  0  5  0]
 [ 0  0  2  0  0  4  0 24 10 10]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  1  1  0  0  4 44]]
Classification report:
               precision    recall  f1-score   support

           0       0.94      0.88      0.91        50
           1       1.00      0.94      0.97        50
           2       0.77      0.94      0.85        50
           3       1.00      0.60      0.75        50
           4       0.97      0.62      0.76        50
           5       0.70      0.74      0.72        50
           6       1.00      0.74      0.85        50
           7       1.00      0.48      0.65        50
           8       0.47      1.00      0.64        50
           9       0.70      0.88      0.78        50

    accuracy                           0.78       500
   macro avg       0.85      0.78      0.79       500
weighted avg       0.85      0.78      0.79       500

Generation time: 1.0829 seconds
Coverage=0.528
Avg set size=0.54
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.505    | 0.64     | 0.64    | 1.16     | 0.51    | 0.20    | 0.12    | 0      
1     | 50   | 0.674    | 0.96     | 0.92    | 1.06     | 0.68    | 0.14    | 0.07    | 1      
2     | 50   | 0.451    | 0.54     | 0.54    | 1.08     | 0.46    | 0.21    | 0.11    | 2      
3     | 50   | 0.334    | 0.28     | 0.28    | 1.78     | 0.39    | 0.23    | 0.16    | 3      
4     | 50   | 0.309    | 0.32     | 0.28    | 1.90     | 0.36    | 0.25    | 0.14    | 4      
5     | 50   | 0.368    | 0.26     | 0.24    | 1.30     | 0.38    | 0.26    | 0.11    | 5      
6     | 50   | 0.423    | 0.48     | 0.48    | 1.62     | 0.45    | 0.18    | 0.13    | 6      
7     | 50   | 0.291    | 0.28     | 0.26    | 2.66     | 0.34    | 0.26    | 0.12    | 7      
8     | 50   | 0.652    | 0.92     | 0.92    | 1.00     | 0.65    | 0.13    | 0.08    | 8      
9     | 50   | 0.502    | 0.72     | 0.72    | 1.24     | 0.51    | 0.17    | 0.11    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.656
Confusion matrix (rows=true, cols=pred):
 [[36  0  0  0  0  7  0  0  7  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 33  0  0  0  0  0 17  0]
 [ 0  0  4 21  0  7  0  0 18  0]
 [ 0  0  1  0 23  1  0  0 17  8]
 [ 0  0  2  0  0 34  0  0 14  0]
 [ 1  0  4  0  0  4 29  0 12  0]
 [ 0  0  2  0  0  5  0 15 17 11]
 [ 0  0  0  0  0  1  0  0 49  0]
 [ 0  0  0  0  0  0  0  0 11 39]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.72      0.83        50
           1       1.00      0.98      0.99        50
           2       0.72      0.66      0.69        50
           3       1.00      0.42      0.59        50
           4       1.00      0.46      0.63        50
           5       0.58      0.68      0.62        50
           6       1.00      0.58      0.73        50
           7       1.00      0.30      0.46        50
           8       0.30      0.98      0.46        50
           9       0.67      0.78      0.72        50

    accuracy                           0.66       500
   macro avg       0.82      0.66      0.67       500
weighted avg       0.82      0.66      0.67       500

Generation time: 1.0630 seconds
Coverage=0.366
Avg set size=0.38
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.374    | 0.30     | 0.30    | 1.60     | 0.39    | 0.22    | 0.15    | 0      
1     | 50   | 0.719    | 0.96     | 0.94    | 1.04     | 0.72    | 0.12    | 0.06    | 1      
2     | 50   | 0.305    | 0.12     | 0.10    | 1.42     | 0.34    | 0.24    | 0.14    | 2      
3     | 50   | 0.271    | 0.16     | 0.16    | 2.30     | 0.34    | 0.24    | 0.17    | 3      
4     | 50   | 0.279    | 0.22     | 0.16    | 2.40     | 0.36    | 0.24    | 0.14    | 4      
5     | 50   | 0.329    | 0.12     | 0.12    | 1.34     | 0.35    | 0.27    | 0.11    | 5      
6     | 50   | 0.311    | 0.28     | 0.28    | 2.00     | 0.36    | 0.19    | 0.15    | 6      
7     | 50   | 0.215    | 0.14     | 0.14    | 3.74     | 0.32    | 0.24    | 0.13    | 8      
8     | 50   | 0.620    | 0.92     | 0.92    | 1.02     | 0.62    | 0.13    | 0.08    | 8      
9     | 50   | 0.444    | 0.54     | 0.54    | 1.42     | 0.47    | 0.17    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9522 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.113    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.049    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.042    | 0.00     | 0.00    | 7.72     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.954
Confusion matrix (rows=true, cols=pred):
 [[46  0  2  0  0  0  1  0  1  0]
 [ 0 49  0  0  0  0  0  0  1  0]
 [ 0  0 49  0  0  0  0  0  1  0]
 [ 0  0  0 49  0  1  0  0  0  0]
 [ 0  0  0  0 45  0  0  0  2  3]
 [ 0  0  1  0  0 45  1  0  3  0]
 [ 0  0  0  0  0  2 48  0  0  0]
 [ 0  0  0  0  0  0  0 47  0  3]
 [ 0  0  1  0  0  0  0  0 49  0]
 [ 0  0  0  0  0  0  0  0  0 50]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.92      0.96        50
           1       1.00      0.98      0.99        50
           2       0.92      0.98      0.95        50
           3       1.00      0.98      0.99        50
           4       1.00      0.90      0.95        50
           5       0.94      0.90      0.92        50
           6       0.96      0.96      0.96        50
           7       1.00      0.94      0.97        50
           8       0.86      0.98      0.92        50
           9       0.89      1.00      0.94        50

    accuracy                           0.95       500
   macro avg       0.96      0.95      0.95       500
weighted avg       0.96      0.95      0.95       500

Generation time: 0.9314 seconds
Coverage=0.892
Avg set size=0.91
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.771    | 0.86     | 0.82    | 1.28     | 0.79    | 0.10    | 0.05    | 0      
1     | 50   | 0.595    | 0.92     | 0.92    | 1.02     | 0.60    | 0.16    | 0.09    | 1      
2     | 50   | 0.890    | 0.92     | 0.92    | 1.02     | 0.89    | 0.05    | 0.02    | 2      
3     | 50   | 0.794    | 0.98     | 0.98    | 1.02     | 0.79    | 0.09    | 0.05    | 3      
4     | 50   | 0.596    | 0.86     | 0.80    | 1.10     | 0.61    | 0.18    | 0.08    | 4      
5     | 50   | 0.621    | 0.82     | 0.78    | 1.10     | 0.63    | 0.19    | 0.08    | 5      
6     | 50   | 0.831    | 1.00     | 0.96    | 1.06     | 0.84    | 0.08    | 0.03    | 6      
7     | 50   | 0.812    | 0.98     | 0.94    | 1.06     | 0.82    | 0.12    | 0.03    | 7      
8     | 50   | 0.691    | 0.92     | 0.92    | 1.04     | 0.69    | 0.14    | 0.07    | 8      
9     | 50   | 0.666    | 0.88     | 0.88    | 1.00     | 0.67    | 0.16    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=4.1806e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.1430 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 8.00     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=1.095e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.926
Confusion matrix (rows=true, cols=pred):
 [[39  0  1  1  0  7  1  0  1  0]
 [ 0 50  0  0  0  0  0  0  0  0]
 [ 0  0 47  1  0  0  0  0  2  0]
 [ 0  0  0 49  0  0  0  0  1  0]
 [ 0  0  0  0 43  0  0  0  0  7]
 [ 0  0  0  2  1 45  0  0  1  1]
 [ 0  0  0  0  0  0 50  0  0  0]
 [ 0  0  1  0  0  0  0 46  0  3]
 [ 0  2  0  1  0  2  0  0 45  0]
 [ 0  0  0  0  0  0  0  1  0 49]]
Classification report:
               precision    recall  f1-score   support

           0       1.00      0.78      0.88        50
           1       0.96      1.00      0.98        50
           2       0.96      0.94      0.95        50
           3       0.91      0.98      0.94        50
           4       0.98      0.86      0.91        50
           5       0.83      0.90      0.87        50
           6       0.98      1.00      0.99        50
           7       0.98      0.92      0.95        50
           8       0.90      0.90      0.90        50
           9       0.82      0.98      0.89        50

    accuracy                           0.93       500
   macro avg       0.93      0.93      0.93       500
weighted avg       0.93      0.93      0.93       500

Generation time: 1.8067 seconds
Coverage=0.916
Avg set size=0.97
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.738    | 0.96     | 0.80    | 1.28     | 0.79    | 0.13    | 0.04    | 0      
1     | 50   | 0.778    | 0.98     | 0.98    | 1.00     | 0.78    | 0.09    | 0.05    | 1      
2     | 50   | 0.903    | 0.98     | 0.94    | 1.08     | 0.91    | 0.04    | 0.02    | 2      
3     | 50   | 0.910    | 0.98     | 0.98    | 1.04     | 0.91    | 0.05    | 0.02    | 3      
4     | 50   | 0.712    | 0.92     | 0.82    | 1.22     | 0.74    | 0.15    | 0.05    | 4      
5     | 50   | 0.704    | 0.92     | 0.86    | 1.14     | 0.74    | 0.15    | 0.05    | 5      
6     | 50   | 0.915    | 1.02     | 1.00    | 1.00     | 0.92    | 0.05    | 0.01    | 6      
7     | 50   | 0.833    | 1.00     | 0.94    | 1.10     | 0.85    | 0.10    | 0.02    | 7      
8     | 50   | 0.723    | 0.92     | 0.88    | 1.12     | 0.74    | 0.14    | 0.06    | 8      
9     | 50   | 0.750    | 1.00     | 0.96    | 1.02     | 0.76    | 0.14    | 0.05    | 9      

  [sigmoid] coef0=0

== KernelPCA-sigmoid(coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.526
Confusion matrix (rows=true, cols=pred):
 [[38  0  0  0  0  6  0  0  6  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 1  0 44  0  0  0  0  0  5  0]
 [ 0  0  1 27  0  1  0  0 21  0]
 [ 0  0  0  0  1  0  0  0 31 18]
 [ 0  0  1  0  0 16  0  0 33  0]
 [ 0  0  6  0  0  0 28  0 16  0]
 [ 0  0  0  0  0  0  0 18  7 25]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0  9 41]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.76      0.85        50
           1       0.00      0.00      0.00        50
           2       0.85      0.88      0.86        50
           3       1.00      0.54      0.70        50
           4       1.00      0.02      0.04        50
           5       0.70      0.32      0.44        50
           6       1.00      0.56      0.72        50
           7       1.00      0.36      0.53        50
           8       0.22      1.00      0.36        50
           9       0.49      0.82      0.61        50

    accuracy                           0.53       500
   macro avg       0.72      0.53      0.51       500
weighted avg       0.72      0.53      0.51       500

Generation time: 1.0945 seconds
Coverage=0.226
Avg set size=0.23
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.431    | 0.44     | 0.44    | 1.42     | 0.45    | 0.21    | 0.14    | 0      
1     | 50   | 0.132    | 0.02     | 0.00    | 2.74     | 0.34    | 0.14    | 0.12    | 8      
2     | 50   | 0.466    | 0.62     | 0.62    | 1.16     | 0.48    | 0.21    | 0.10    | 2      
3     | 50   | 0.297    | 0.12     | 0.10    | 1.60     | 0.34    | 0.25    | 0.17    | 3      
4     | 50   | 0.161    | 0.00     | 0.00    | 3.02     | 0.27    | 0.21    | 0.16    | 8      
5     | 50   | 0.292    | 0.02     | 0.02    | 1.72     | 0.35    | 0.27    | 0.11    | 8      
6     | 50   | 0.284    | 0.10     | 0.10    | 1.68     | 0.32    | 0.22    | 0.16    | 6      
7     | 50   | 0.279    | 0.06     | 0.06    | 1.90     | 0.32    | 0.26    | 0.15    | 9      
8     | 50   | 0.485    | 0.82     | 0.82    | 1.00     | 0.48    | 0.16    | 0.11    | 8      
9     | 50   | 0.331    | 0.10     | 0.10    | 1.18     | 0.35    | 0.22    | 0.13    | 9      

  [sigmoid] coef0=1

== KernelPCA-sigmoid(coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.0342 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.070    | 0.00     | 0.00    | 5.40     | 0.32    | 0.20    | 0.12    | 8      
1     | 50   | 0.039    | 0.00     | 0.00    | 8.66     | 0.35    | 0.14    | 0.11    | 8      
2     | 50   | 0.162    | 0.00     | 0.00    | 2.28     | 0.34    | 0.17    | 0.14    | 8      
3     | 50   | 0.116    | 0.00     | 0.00    | 3.58     | 0.34    | 0.17    | 0.12    | 8      
4     | 50   | 0.057    | 0.00     | 0.00    | 6.10     | 0.33    | 0.15    | 0.13    | 8      
5     | 50   | 0.183    | 0.00     | 0.00    | 2.00     | 0.35    | 0.18    | 0.11    | 8      
6     | 50   | 0.071    | 0.00     | 0.00    | 5.68     | 0.32    | 0.16    | 0.13    | 8      
7     | 50   | 0.066    | 0.00     | 0.00    | 5.92     | 0.32    | 0.15    | 0.14    | 8      
8     | 50   | 0.374    | 0.00     | 0.00    | 1.00     | 0.37    | 0.16    | 0.11    | 8      
9     | 50   | 0.141    | 0.00     | 0.00    | 2.68     | 0.34    | 0.15    | 0.14    | 8      


--- latent_size=12 | std=2.0 ---

== PCA [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.592
Confusion matrix (rows=true, cols=pred):
 [[25  0  8  1  0 10  1  1  3  1]
 [ 0 35  1  4  0  1  0  1  8  0]
 [ 2  1 27  3  0 11  2  1  2  1]
 [ 3  2 11 24  0  6  0  0  4  0]
 [ 0  0  1  1 29  5  0  4  5  5]
 [ 1  0  6  6  4 28  0  0  5  0]
 [ 0  1  3  0  4  5 35  0  1  1]
 [ 0  1  4  0  1  0  0 37  0  7]
 [ 0  1  9  3  1  7  3  0 23  3]
 [ 0  0  2  0  0  3  1  5  6 33]]
Classification report:
               precision    recall  f1-score   support

           0       0.81      0.50      0.62        50
           1       0.85      0.70      0.77        50
           2       0.38      0.54      0.44        50
           3       0.57      0.48      0.52        50
           4       0.74      0.58      0.65        50
           5       0.37      0.56      0.44        50
           6       0.83      0.70      0.76        50
           7       0.76      0.74      0.75        50
           8       0.40      0.46      0.43        50
           9       0.65      0.66      0.65        50

    accuracy                           0.59       500
   macro avg       0.64      0.59      0.60       500
weighted avg       0.64      0.59      0.60       500

Generation time: 0.6831 seconds
Coverage=0.536
Avg set size=0.81
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.366    | 0.80     | 0.42    | 2.20     | 0.56    | 0.23    | 0.10    | 0      
1     | 50   | 0.514    | 0.82     | 0.60    | 1.46     | 0.58    | 0.19    | 0.10    | 1      
2     | 50   | 0.459    | 0.80     | 0.50    | 2.60     | 0.65    | 0.16    | 0.08    | 2      
3     | 50   | 0.437    | 0.80     | 0.48    | 2.24     | 0.59    | 0.20    | 0.09    | 3      
4     | 50   | 0.471    | 0.74     | 0.52    | 1.86     | 0.62    | 0.17    | 0.09    | 4      
5     | 50   | 0.473    | 0.84     | 0.50    | 1.84     | 0.60    | 0.20    | 0.09    | 5      
6     | 50   | 0.537    | 0.80     | 0.66    | 1.54     | 0.61    | 0.19    | 0.08    | 6      
7     | 50   | 0.583    | 0.90     | 0.70    | 1.32     | 0.65    | 0.19    | 0.07    | 7      
8     | 50   | 0.382    | 0.78     | 0.38    | 1.84     | 0.56    | 0.23    | 0.10    | 8      
9     | 50   | 0.489    | 0.84     | 0.60    | 1.64     | 0.60    | 0.20    | 0.08    | 9      

  [rbf] gamma=0.002

== KernelPCA-rbf(gamma=0.002) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.702
Confusion matrix (rows=true, cols=pred):
 [[39  0  2  1  0  6  0  0  2  0]
 [ 0 42  0  0  0  0  0  0  8  0]
 [ 0  0 46  0  0  0  0  0  4  0]
 [ 0  0  6 24  0  8  0  0 12  0]
 [ 1  0  1  0 21  1  0  0 14 12]
 [ 0  0  4  0  1 31  0  0 14  0]
 [ 1  0  7  0  0  2 31  0  9  0]
 [ 0  0  6  0  0  6  0 27  6  5]
 [ 0  0  1  0  0  1  0  0 47  1]
 [ 0  0  1  0  0  1  0  0  5 43]]
Classification report:
               precision    recall  f1-score   support

           0       0.95      0.78      0.86        50
           1       1.00      0.84      0.91        50
           2       0.62      0.92      0.74        50
           3       0.96      0.48      0.64        50
           4       0.95      0.42      0.58        50
           5       0.55      0.62      0.58        50
           6       1.00      0.62      0.77        50
           7       1.00      0.54      0.70        50
           8       0.39      0.94      0.55        50
           9       0.70      0.86      0.77        50

    accuracy                           0.70       500
   macro avg       0.81      0.70      0.71       500
weighted avg       0.81      0.70      0.71       500

Generation time: 1.0709 seconds
Coverage=0.464
Avg set size=0.50
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.489    | 0.56     | 0.56    | 1.46     | 0.51    | 0.20    | 0.12    | 0      
1     | 50   | 0.509    | 0.66     | 0.62    | 1.20     | 0.53    | 0.20    | 0.10    | 1      
2     | 50   | 0.485    | 0.66     | 0.66    | 1.12     | 0.49    | 0.18    | 0.11    | 2      
3     | 50   | 0.321    | 0.34     | 0.22    | 2.04     | 0.41    | 0.23    | 0.15    | 3      
4     | 50   | 0.265    | 0.24     | 0.18    | 2.40     | 0.34    | 0.24    | 0.14    | 4      
5     | 50   | 0.347    | 0.34     | 0.30    | 1.44     | 0.38    | 0.25    | 0.13    | 5      
6     | 50   | 0.379    | 0.38     | 0.34    | 1.80     | 0.44    | 0.19    | 0.12    | 6      
7     | 50   | 0.285    | 0.38     | 0.30    | 2.96     | 0.37    | 0.26    | 0.12    | 7      
8     | 50   | 0.590    | 0.86     | 0.86    | 1.06     | 0.59    | 0.15    | 0.09    | 8      
9     | 50   | 0.468    | 0.60     | 0.60    | 1.40     | 0.49    | 0.17    | 0.12    | 9      

  [rbf] gamma=0.003

== KernelPCA-rbf(gamma=0.003) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.718
Confusion matrix (rows=true, cols=pred):
 [[39  0  2  1  0  7  0  0  1  0]
 [ 0 45  0  0  0  0  0  0  5  0]
 [ 0  0 40  0  0  1  0  0  9  0]
 [ 0  0  7 26  0  6  0  0 11  0]
 [ 2  0  2  1 26  1  1  0  8  9]
 [ 0  0  5  0  0 34  0  0 11  0]
 [ 2  0  3  0  0  2 36  0  7  0]
 [ 0  0  5  0  0  6  0 23  7  9]
 [ 0  0  0  0  0  2  0  0 47  1]
 [ 0  0  0  0  0  2  1  0  4 43]]
Classification report:
               precision    recall  f1-score   support

           0       0.91      0.78      0.84        50
           1       1.00      0.90      0.95        50
           2       0.62      0.80      0.70        50
           3       0.93      0.52      0.67        50
           4       1.00      0.52      0.68        50
           5       0.56      0.68      0.61        50
           6       0.95      0.72      0.82        50
           7       1.00      0.46      0.63        50
           8       0.43      0.94      0.59        50
           9       0.69      0.86      0.77        50

    accuracy                           0.72       500
   macro avg       0.81      0.72      0.73       500
weighted avg       0.81      0.72      0.73       500

Generation time: 1.2306 seconds
Coverage=0.512
Avg set size=0.55
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.492    | 0.60     | 0.60    | 1.48     | 0.51    | 0.19    | 0.12    | 0      
1     | 50   | 0.632    | 0.86     | 0.80    | 1.16     | 0.66    | 0.15    | 0.08    | 1      
2     | 50   | 0.446    | 0.54     | 0.54    | 1.22     | 0.46    | 0.20    | 0.12    | 2      
3     | 50   | 0.332    | 0.42     | 0.30    | 2.02     | 0.42    | 0.22    | 0.15    | 3      
4     | 50   | 0.299    | 0.36     | 0.28    | 2.48     | 0.38    | 0.24    | 0.13    | 4      
5     | 50   | 0.350    | 0.30     | 0.26    | 1.40     | 0.39    | 0.25    | 0.12    | 5      
6     | 50   | 0.425    | 0.54     | 0.48    | 1.84     | 0.48    | 0.18    | 0.12    | 6      
7     | 50   | 0.292    | 0.34     | 0.30    | 3.04     | 0.37    | 0.26    | 0.11    | 7      
8     | 50   | 0.634    | 0.92     | 0.90    | 1.08     | 0.64    | 0.14    | 0.08    | 8      
9     | 50   | 0.485    | 0.66     | 0.66    | 1.48     | 0.51    | 0.18    | 0.11    | 9      

  [rbf] gamma=0.005

== KernelPCA-rbf(gamma=0.005) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.628
Confusion matrix (rows=true, cols=pred):
 [[33  0  1  0  0  8  0  0  8  0]
 [ 0 45  0  0  0  0  0  0  5  0]
 [ 0  0 31  0  0  3  0  0 16  0]
 [ 0  0  5 21  0  8  0  0 16  0]
 [ 0  0  3  0 23  2  0  0 14  8]
 [ 0  0  3  0  1 31  0  0 15  0]
 [ 1  0  5  0  0  4 28  0 12  0]
 [ 0  0  5  0  0  5  0 17 15  8]
 [ 0  0  0  0  0  2  0  0 48  0]
 [ 0  0  1  0  0  2  0  0 10 37]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.66      0.79        50
           1       1.00      0.90      0.95        50
           2       0.57      0.62      0.60        50
           3       1.00      0.42      0.59        50
           4       0.96      0.46      0.62        50
           5       0.48      0.62      0.54        50
           6       1.00      0.56      0.72        50
           7       1.00      0.34      0.51        50
           8       0.30      0.96      0.46        50
           9       0.70      0.74      0.72        50

    accuracy                           0.63       500
   macro avg       0.80      0.63      0.65       500
weighted avg       0.80      0.63      0.65       500

Generation time: 1.0403 seconds
Coverage=0.374
Avg set size=0.41
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.377    | 0.32     | 0.32    | 1.82     | 0.42    | 0.21    | 0.14    | 0      
1     | 50   | 0.658    | 0.86     | 0.78    | 1.14     | 0.68    | 0.15    | 0.07    | 1      
2     | 50   | 0.304    | 0.18     | 0.16    | 1.56     | 0.35    | 0.24    | 0.13    | 2      
3     | 50   | 0.278    | 0.26     | 0.20    | 2.56     | 0.37    | 0.23    | 0.16    | 3      
4     | 50   | 0.276    | 0.34     | 0.22    | 2.90     | 0.38    | 0.23    | 0.13    | 4      
5     | 50   | 0.318    | 0.20     | 0.16    | 1.44     | 0.35    | 0.25    | 0.12    | 5      
6     | 50   | 0.320    | 0.36     | 0.32    | 2.30     | 0.40    | 0.18    | 0.14    | 6      
7     | 50   | 0.222    | 0.20     | 0.20    | 4.12     | 0.35    | 0.24    | 0.13    | 7      
8     | 50   | 0.607    | 0.86     | 0.86    | 1.04     | 0.61    | 0.14    | 0.08    | 8      
9     | 50   | 0.428    | 0.52     | 0.52    | 1.88     | 0.47    | 0.17    | 0.12    | 9      

  [poly] degree=2 coef0=0

== KernelPCA-poly(degree=2,coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 0.9328 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.114    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.089    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.043    | 0.00     | 0.00    | 6.98     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.162    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.050    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.043    | 0.00     | 0.00    | 7.46     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=2 coef0=1

== KernelPCA-poly(degree=2,coef0=1) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.880
Confusion matrix (rows=true, cols=pred):
 [[38  0  6  0  2  2  0  0  2  0]
 [ 0 48  1  0  0  0  0  0  1  0]
 [ 1  1 45  0  0  1  0  0  2  0]
 [ 0  0  1 48  0  1  0  0  0  0]
 [ 0  0  1  0 41  0  0  1  3  4]
 [ 0  0  3  2  1 38  1  0  4  1]
 [ 0  0  0  0  0  2 48  0  0  0]
 [ 0  0  0  0  0  0  0 45  1  4]
 [ 0  2  1  0  0  1  1  0 45  0]
 [ 0  0  0  3  0  1  0  2  0 44]]
Classification report:
               precision    recall  f1-score   support

           0       0.97      0.76      0.85        50
           1       0.94      0.96      0.95        50
           2       0.78      0.90      0.83        50
           3       0.91      0.96      0.93        50
           4       0.93      0.82      0.87        50
           5       0.83      0.76      0.79        50
           6       0.96      0.96      0.96        50
           7       0.94      0.90      0.92        50
           8       0.78      0.90      0.83        50
           9       0.83      0.88      0.85        50

    accuracy                           0.88       500
   macro avg       0.89      0.88      0.88       500
weighted avg       0.89      0.88      0.88       500

Generation time: 0.9415 seconds
Coverage=0.830
Avg set size=0.89
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.708    | 0.92     | 0.76    | 1.76     | 0.77    | 0.12    | 0.05    | 0      
1     | 50   | 0.587    | 0.90     | 0.88    | 1.06     | 0.59    | 0.16    | 0.09    | 1      
2     | 50   | 0.861    | 0.94     | 0.90    | 1.18     | 0.88    | 0.05    | 0.02    | 2      
3     | 50   | 0.779    | 0.94     | 0.92    | 1.08     | 0.78    | 0.10    | 0.05    | 3      
4     | 50   | 0.572    | 0.82     | 0.74    | 1.32     | 0.61    | 0.18    | 0.08    | 4      
5     | 50   | 0.600    | 0.80     | 0.70    | 1.26     | 0.64    | 0.18    | 0.08    | 5      
6     | 50   | 0.797    | 0.90     | 0.86    | 1.10     | 0.82    | 0.10    | 0.03    | 6      
7     | 50   | 0.796    | 0.92     | 0.86    | 1.12     | 0.81    | 0.11    | 0.04    | 7      
8     | 50   | 0.640    | 0.88     | 0.84    | 1.20     | 0.66    | 0.16    | 0.08    | 8      
9     | 50   | 0.639    | 0.84     | 0.84    | 1.20     | 0.66    | 0.17    | 0.08    | 9      

  [poly] degree=3 coef0=0
/usr/local/lib/python3.11/dist-packages/sklearn/decomposition/_kernel_pca.py:410: LinAlgWarning: Ill-conditioned matrix (rcond=4.1806e-08): result may not be accurate.
  self.dual_coef_ = linalg.solve(K, X, assume_a="pos", overwrite_a=True)

== KernelPCA-poly(degree=3,coef0=0) [Logistic] | latent=12² ==
Classifier accuracy (pointwise): 0.100
Confusion matrix (rows=true, cols=pred):
 [[ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]
 [ 0  0  0  0  0  0  0  0 50  0]]
Classification report:
               precision    recall  f1-score   support

           0       0.00      0.00      0.00        50
           1       0.00      0.00      0.00        50
           2       0.00      0.00      0.00        50
           3       0.00      0.00      0.00        50
           4       0.00      0.00      0.00        50
           5       0.00      0.00      0.00        50
           6       0.00      0.00      0.00        50
           7       0.00      0.00      0.00        50
           8       0.10      1.00      0.18        50
           9       0.00      0.00      0.00        50

    accuracy                           0.10       500
   macro avg       0.01      0.10      0.02       500
weighted avg       0.01      0.10      0.02       500

Generation time: 1.1498 seconds
Coverage=0.000
Avg set size=0.00
Digit | N    | Conf     | SetSize  | Covrg   | TrueRank | Top1    | Top2    | Top3    | #TopPred
------------------------------------------------------------------------------------------------
0     | 50   | 0.036    | 0.00     | 0.00    | 9.00     | 0.34    | 0.16    | 0.11    | 8      
1     | 50   | 0.027    | 0.00     | 0.00    | 10.00    | 0.34    | 0.16    | 0.11    | 8      
2     | 50   | 0.112    | 0.00     | 0.00    | 3.00     | 0.34    | 0.16    | 0.11    | 8      
3     | 50   | 0.088    | 0.00     | 0.00    | 5.00     | 0.34    | 0.16    | 0.11    | 8      
4     | 50   | 0.042    | 0.00     | 0.00    | 7.00     | 0.34    | 0.16    | 0.11    | 8      
5     | 50   | 0.161    | 0.00     | 0.00    | 2.00     | 0.34    | 0.16    | 0.11    | 8      
6     | 50   | 0.048    | 0.00     | 0.00    | 6.00     | 0.34    | 0.16    | 0.11    | 8      
7     | 50   | 0.041    | 0.00     | 0.00    | 7.94     | 0.34    | 0.16    | 0.11    | 8      
8     | 50   | 0.343    | 0.00     | 0.00    | 1.00     | 0.34    | 0.16    | 0.11    | 8      
9     | 50   | 0.102    | 0.00     | 0.00    | 4.00     | 0.34    | 0.16    | 0.11    | 8      

  [poly] degree=3 coef0=1