lightning/pytorch_lightning/trainer
William Falcon 00f0d19a61
fixes #3798 (#3849)
* fix #3798

* added tbptt test for logging
2020-10-04 19:36:51 -04:00
..
connectors test selecting the correct backend. temp backends while slurm and TE are decoupled (#3848) 2020-10-04 15:44:50 -04:00
__init__.py fixes #3798 (#3849) 2020-10-04 19:36:51 -04:00
callback_hook.py fix on_fit_start (#3616) 2020-09-23 10:38:33 +02:00
configuration_validator.py ref: added data connector (#3285) 2020-08-31 11:08:22 -04:00
data_loading.py revert backend types (#3788) 2020-10-02 06:18:44 -04:00
deprecated_api.py drop v0.10 deprecated (#3454) 2020-09-19 11:47:26 -04:00
evaluation_loop.py fixes #3798 (#3849) 2020-10-04 19:36:51 -04:00
ignored_warnings.py
logging.py Use `raise .. from ..` to explicitly chain exceptions (#3750) 2020-10-01 21:45:44 +02:00
model_hooks.py ref: modular is_overridden (#3290) 2020-08-31 12:12:02 -04:00
optimizers.py ref: bug fix with logging val epoch end + monitor (#3812) 2020-10-03 12:33:29 -04:00
properties.py ref: slurm connector 1/n (#3476) 2020-09-12 11:07:15 -04:00
states.py Make trainer.state a read-only property (#3109) 2020-08-24 16:49:33 +02:00
supporters.py
trainer.py Eval epoch can now log independently (#3843) 2020-10-04 13:36:35 -04:00
training_loop.py ref: bug fix with logging val epoch end + monitor (#3812) 2020-10-03 12:33:29 -04:00
training_tricks.py ref: checkpoint connector methods 4/n (#3474) 2020-09-12 08:42:27 -04:00