table of Contents
"Structured Machine Learning Project" | Note list |
---|---|
Week 1 | Machine Learning Strategy One |
Week 1 Portal—> | [1.1 Why is ML Strategy] [1.2 Orthogonalization] [1.3 Single Digital Evaluation Index] [1.4 Satisfaction and Optimization Index] [1.5 Training / Development / Test Set Division] [1.6 Development Set Test Set Size] [1.7 When The change development _ test set and indicators] [1.8 Why is human performance] [1.9 avoidable errors] [1.10 understand human performance] [1.11 beyond human performance] [1.12 improve your model ’s performance] |
Week 2 | Machine Learning Strategy 2 |
Week 2 Portal—> | [2.1 Error analysis] [2.2 Clear incorrectly marked data] [2.3 Quickly build your first system and iterate] [2.4 Train and test on different partitions] [2.5 Deviations and errors of mismatched data partitions ] [2.6 Location data mismatch] [2.7 Transfer learning] [2.8 Multi-task learning] [2.9 What is end-to-end deep learning] [2.10 Whether to use end-to-end deep learning] |
Interview | Daniel interview |
Portal —> | [Andrej Karpathy] [Ruslan Salakhutdinov] |