Adrian Wälchli 1e68968ed7 support num_sanity_val_steps=-1 (#2246) ... * support sanity_val_step=-1 * fix list size * simplification * simplify * add test for num_sanity_val_steps=-1 * update test * update docs * extend tests to multiple dataloaders * changelog * Update tests/trainer/test_trainer.py Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com> * improve test * refactor the sanity check decision * fix merge * Update trainer.py Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com> Co-authored-by: William Falcon <waf2107@columbia.edu>		2020-07-23 07:07:03 -04:00
..
callbacks	support num_sanity_val_steps=-1 (#2246)	2020-07-23 07:07:03 -04:00
core	EvalResult support for val loop (PR 3/5) (#2651)	2020-07-22 13:53:10 -04:00
loggers	save_dir fix for MLflowLogger + save_dir tests for others (#2502)	2020-07-09 07:15:41 -04:00
logging	remove deprecated API for v0.8 (#2073)	2020-06-12 14:37:52 -04:00
metrics	metrics: add BLEU (#2535)	2020-07-22 09:58:24 -04:00
overrides	Structured results (train loop only. val loop separate PR) (PR 2/5) (#2615)	2020-07-20 19:00:20 -04:00
profiler	replace Hparams by init args (#1896)	2020-05-24 18:59:08 -04:00
trainer	support num_sanity_val_steps=-1 (#2246)	2020-07-23 07:07:03 -04:00
utilities	EvalResult support for val loop (PR 3/5) (#2651)	2020-07-22 13:53:10 -04:00
__init__.py	Structured results (train loop only. val loop separate PR) (PR 2/5) (#2615)	2020-07-20 19:00:20 -04:00