Menu

trainlint: static checks for ML bugs that don't crash (fit-before-split, eval mode, logits losses) [P]

Reading 0:00
15s threshold

I wrote a small checker for training and evaluation code. It looks for mistakes that do not raise errors. The classic one is leakage through preprocessing.…

Anonymous readers can preview up to 1024 characters here. Log in to unlock the full article once ingest succeeds.