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
..
__init__.py	support num_sanity_val_steps=-1 (#2246)	2020-07-23 07:07:03 -04:00
auto_mix_precision.py	native amp (#2373)	2020-06-26 21:45:13 -04:00
callback_config.py	Continue Jeremy's early stopping PR #1504 (#2391)	2020-06-28 21:36:46 -04:00
callback_hook.py	Structured results (train loop only. val loop separate PR) (PR 2/5) (#2615)	2020-07-20 19:00:20 -04:00
data_loading.py	Fix ddp tests + .test() (#2512)	2020-07-07 12:24:56 -04:00
deprecated_api.py	deprecated Trainer proc_rank (#2269)	2020-06-19 15:46:27 -04:00
distrib_data_parallel.py	Fixes .test() for ddp (#2570)	2020-07-09 18:36:36 -04:00
distrib_parts.py	Ampt (#2572)	2020-07-09 21:28:11 -04:00
evaluation_loop.py	support num_sanity_val_steps=-1 (#2246)	2020-07-23 07:07:03 -04:00
ignored_warnings.py	Dim 0 warning (#256)	2019-09-26 13:20:54 -04:00
logging.py	Structured results (train loop only. val loop separate PR) (PR 2/5) (#2615)	2020-07-20 19:00:20 -04:00
lr_finder.py	Checking ipywidgets is installed for ensure tqdm working (#2417)	2020-06-30 16:59:35 -04:00
model_hooks.py	EvalResult support for val loop (PR 3/5) (#2651)	2020-07-22 13:53:10 -04:00
optimizers.py	swaps lr sched order (#2356)	2020-06-25 09:21:41 -04:00
supporters.py	Structured results (train loop only. val loop separate PR) (PR 2/5) (#2615)	2020-07-20 19:00:20 -04:00
trainer.py	support num_sanity_val_steps=-1 (#2246)	2020-07-23 07:07:03 -04:00
training_io.py	Ampt (#2572)	2020-07-09 21:28:11 -04:00
training_loop.py	support num_sanity_val_steps=-1 (#2246)	2020-07-23 07:07:03 -04:00
training_tricks.py	Start accumulate gradients schedule at epoch 0 (continued) (#2513)	2020-07-09 07:11:07 -04:00