ML 5.2Lesson outlineLesson 2 of 5 in Module 5
Thresholds, ROC and AUC
Treat a classifier's output as a score, read the trade every threshold makes, and compare models across all thresholds at once.
Video in production.
Lesson outline: Thresholds, ROC and AUC
In this lesson9 parts
- 01Four votes out of seven
- 02Eight possible scores
- 03Comparing by accuracy at one threshold
- 04One threshold, then the next, on ten tumors
- 05The area, and the pairs
- 06Reading the AUC
- 07What the AUC does not measure
- 08Code
- 09Recap, your challenge, and what comes next
Builds on
This lesson stands on its own. These go deeper into what it uses.