wassname b2c666dbbf add wassname's 37-reasons checklist to refs/checklist.md ...

Quoted whole from the 2017 thread, since the article author asked to merge it
and never did. Verified line by line against the thread cache. Covers sample
size from a cumulative-mean plot, KLD/Dice on unbalanced data, augmentation
bounded by feature std, dummy metrics, jumpy validation loss, testing the
framework itself, activation swaps, and loss-curve shapes.

2026-08-15 06:40:45 +08:00

..

evidence

add wassname's 37-reasons checklist to refs/checklist.md

2026-08-15 06:40:45 +08:00

spec

Record ML debugging benchmark results

2026-07-13 06:11:08 +08:00

ml_debug_folklore.argdown

initial: ML debugging folklore skill

2026-03-06 10:11:30 +08:00