ludwig
ludwig copied to clipboard
AutoML: stratify imbalanced datasets
Adds stratified splitting to the AutoML generated config if the dataset:
- has a single output feature
- the output feature is categorical or binary
- the output feature's imbalance ratio is below
imbalance_threshold
Unit Test Results
6 files + 1 6 suites +1 3h 36m 47s :stopwatch: + 1h 16m 17s 3 415 tests + 2 3 337 :heavy_check_mark: + 2 78 :zzz: ± 0 0 :x: ±0 10 245 runs +140 9 988 :heavy_check_mark: +116 257 :zzz: +24 0 :x: ±0
Results for commit 6b7f9f98. ± Comparison against base commit 1dc66cad.
:recycle: This comment has been updated with latest results.