Classification Section : Multi-class classification part 7: Evaluating our model (8:32) #25
-
|
Daniel I tf.one_hot() encoded the test and train labels when fitting the model that seems to work for me I passed the normal train_labels and I got the probs. I don't know if this is correct. |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 5 replies
-
|
Woah! That's definitely one option as well. It also depends on what loss function you use (sparse categorical crossentropy vs categorical crossentropy). Did your results look okay? |
Beta Was this translation helpful? Give feedback.
-
|
Yes @mrdbourke :) |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for the reponse |
Beta Was this translation helpful? Give feedback.
-
|
The confusion matrix looks different |
Beta Was this translation helpful? Give feedback.
-
|
Oh really? What's the main difference? Did it make sense? |
Beta Was this translation helpful? Give feedback.
-
|
all of the values which were supposed to go from the left to right almost all of were in one corner. |
Beta Was this translation helpful? Give feedback.
Woah! That's definitely one option as well. It also depends on what loss function you use (sparse categorical crossentropy vs categorical crossentropy).
Did your results look okay?