cap done

vishnoitanuj · Nov 8, 2018 · ee3a137 · ee3a137
1 parent 2df765d
commit ee3a137
Show file tree

Hide file tree

Showing 3 changed files with 11 additions and 0 deletions.
diff --git a/...ssification/Section 21 - Evaluating Classification Models Performance/README.md b/...ssification/Section 21 - Evaluating Classification Models Performance/README.md
@@ -2,3 +2,14 @@
 
 1) False Positive(Type 1 error): We predicted a positive outcome but it was false. Predicted an effet that did not occur. [Less Dangerous]
 2) False Negative(Type 2 error): We predicted that there won't be any effect but it actually did occur. Our prediction was negative. [More Dangerous]
+
+# Confusion Matrix
+![](cm.png)
+
+###### Accuracy Paradox:
+If 00 = more, 01 = 0, 01 = less, 11 = 0
+Accuracy increases, although we did not train. So accuarcy cannot be always trusted.
+
+#CAP: Cummulative Accuracy Profile.
+Greater area under the curve, the better is the model.  
+![](cap.png)
diff --git a/...lassification/Section 21 - Evaluating Classification Models Performance/cap.png b/...lassification/Section 21 - Evaluating Classification Models Performance/cap.png
diff --git a/...Classification/Section 21 - Evaluating Classification Models Performance/cm.png b/...Classification/Section 21 - Evaluating Classification Models Performance/cm.png