Sweep max_depth over [1, 2, 3, 5, 8, 12] using 5-fold roc_auc, and return the d…
Machine Learning · from Overfitting and underfitting
bias-variancevalidation_curvelearning_curvemax_depthtrain vs test gapPutting the kettle on…
Starting up…
Exercise 3
Sweep max_depth over [1, 2, 3, 5, 8, 12] using 5-fold roc_auc, and return the depth with the best cross-validated score — not the best training score.
your answer
stuck? the lesson behind it
Overfitting and underfitting
Two failures that look identical from a bad score, with opposite cures