summary of Structuring Machine Learning Projects

Note

This is my personal summary after studying the course, Structuring Machine Learning Projects and the copyright belongs to deeplearning.ai.

Content Table of my personal notes

$1_{st}$ week: 01_ml-strategy-1

• 01_introduction-to-ml-strategy
  • 01_why-ml-strategy
  • 02_orthogonalization
    • Orthogonalization
• 02_setting-up-your-goal
  • 01_single-number-evaluation-metric
  • 02_satisficing-and-optimizing-metric
    • summary
  • 03_train-dev-test-distributions
    • summary
  • 04_size-of-the-dev-and-test-sets
    • Summary
  • 05_when-to-change-dev-test-sets-and-metrics
    • summary
• 03_comparing-to-human-level-performance
  • 01_why-human-level-performance
    • summary
  • 02_avoidable-bias
    • summary
  • 03_understanding-human-level-performance
    • Summary
• 04_surpassing-human-level-performance
  • summary
• 05_improving-your-model-performance

$2_{nd}$ week: 02_ml-strategy-2

• 01_error-analysis
  • 01_carrying-out-error-analysis
  • 02_cleaning-up-incorrectly-labeled-data
  • 03_build-your-first-system-quickly-then-iterate
• 02_mismatched-training-and-dev-test-set
  • 01_training-and-testing-on-different-distributions
  • 02_bias-and-variance-with-mismatched-data-distributions
  • 03_addressing-data-mismatch
• 03_learning-from-multiple-tasks
  • 01_transfer-learning
  • 02_multi-task-learning
• 04_end-to-end-deep-learning
  • 01_what-is-end-to-end-deep-learning
  • 02_whether-to-use-end-to-end-deep-learning