Anand Avati
Computer Science, PhD
To follow along with the course schedule and syllabus, visit:
0:00 Introduction
0:42 Topics
1:16 Why are metrics important?
5:23 Binary Classification
11:55 Score based models : Classifier
13:40 Point metrics: Confusion Matrix
15:05 Point metrics: True Positives
15:56 Point metrics: True Negatives
16:28 Point metrics: False Positives
16:49 Point metrics: False Negatives
17:43 FP and FN also called Type-1 and Type-2 errors
19:38 Point metrics: Accuracy
20:29 Point metrics: Precision
23:15 Point metrics: Positive Recall (Sensitivity)
28:22 Point metrics: Negative Recall (Specificity)
33:48 Point metrics: F score
35:40 Point metrics: Changing threshold
46:05 Summary metrics: ROC (rotated version)
54:02 Summary metrics: PRC
59:05 Summary metrics: Log-Loss motivation