Lecture 03: Troubleshooting & Testing (FSDL 2022)

In this video, we cover what you need to test ML codebases and troubleshoot deep neural networks.

00:00 Overview
00:33 Testing software
05:05 Testing tools: pytest, doctests, codecov
08:00 Clean code tools: black, flake8, shellcheck
12:00 Automation
13:43 GitHub Actions for automation
16:58 Testing ML systems
19:07 Testing data
22:21 Testing training
26:43 Testing models
30:59 Test in production
32:33 The ML Test Score
34:05 Troubleshooting models
37:56 Troubleshooting performance
41:47 Outro

Subscribe to our channel and sign up at …

testing troubleshooting