Week 2: Machine Learning Strategy and Best Practices
Lecture
Note
Status: Lecture notes and slides are available.
Overview
This lecture covers how to think about machine learning projects: ML strategy and orthogonalization, choosing evaluation metrics and data splits, diagnosing bias, variance, and data mismatch, prioritizing work with error analysis, and a short conceptual introduction to serving a model with an API and Docker.
Materials
| Material | Where |
|---|---|
| Lecture notes | read |
| Slides | open (view in browser) |
| Lab | - |
| Papers | - |
References
[1] A. Ng, Deep Learning Specialization, DeepLearning.AI.
[2] FastAPI documentation, https://fastapi.tiangolo.com
[3] Docker documentation, https://docs.docker.com