Building a Visualizer in Yellowbrick
summary
Yellowbrick extends the scikit-learn API to produce matplotlib plots glued with numpy and packaged as a class with init fit draw poof lifecycle. Workflow shows building a decision‑boundary visualizer for k‑nearest and random forest, writing visual unit tests that use sklearn make_classification with a fixed random seed, mocking matplotlib where needed, and relying on an image comparison test against baseline PNGs to assert stability of the fit draw poof API.
AI-written summary · about
chapters (8)
- 0:00Yellowbrick introduction
- 10:32Yellowbrick visualizers overview
- 16:18Develop visualizer step-by-step
- 21:17Refactor code to Yellowbrick API
- 26:43Test and document visualizers
- 33:11Generate test data
- 35:17Contributing best practices
- 36:49Q&A
related talks
39mYellowbrick: Steering Machine Learning with Visual Transformers
39mVisual diagnostics for more informed machine learning
35mVisual Diagnostics for More Effective Machine Learning
43mVisualising Machine Learning Models
30mWords in Space
38mBuilding a maintainable plotting library-PyData NYC
43mAdvanced Data Visualizations In Jupyter Notebooks
01h 38mEffective Data Visualization