Confusion Matrix

The building block of all classification metrics.

Definitions

- **TP** (True Positive): predicted 1, actually 1

- **FP** (False Positive): predicted 1, actually 0

- **FN** (False Negative): predicted 0, actually 1

- **TN** (True Negative): predicted 0, actually 0

Example

confusion_matrix([1,0,1,1,0,1],[1,0,0,1,1,1])
→ {'TP': 3, 'FP': 1, 'FN': 1, 'TN': 1}