Completed
Key points for choosing a metric
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Setting Up Machine Learning Projects - Full Stack Deep Learning - March 2019
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Goals for the lecture
- 3 Running case study - pose estimation
- 4 Hypothetical Co. Full Stack Robotics (FSR) wants to use pose estimation to enable grasping
- 5 Lifecycle of a ML project
- 6 Outline of the rest of the lecture
- 7 Key points for prioritizing projects
- 8 A (general) framework for prioritizing projects
- 9 Why are accuracy requirements so important?
- 10 Product design can reduce need for accuracy
- 11 Another heuristic for assessing feasibility
- 12 Key points for choosing a metric
- 13 Review of accuracy, precision, and recall
- 14 Why choose a single metric?
- 15 How to combine metrics
- 16 Combining precision and recall
- 17 Thresholding metrics
- 18 Example: choosing a metric for pose estimation
- 19 How to create good human baselines Quality of baseline Low
- 20 Key points for choosing baselines
- 21 Questions?
