Top pytorch-lightning issues

TensorBoardLogger log_hyperparameters fails when explicitly specified.

bug / fix help wanted
BrianPugh picture BrianPugh·22 Sep 2020·9 Comments

Infinite hang when running `Trainer.test` after `Trainer.fit` with DDP

Working as intended bug / fix duplicate help wanted
josh-gleason picture josh-gleason·22 Sep 2020·8 Comments

How to access validation logs data inside the on_validation_batch_end callback

question
DomainFlag picture DomainFlag·21 Sep 2020·11 Comments

Incorrect "Saving latest checkpoint" warning

Checkpoint bug / fix help wanted
carmocca picture carmocca·21 Sep 2020·5 Comments

Enabling batch transforms

documentation enhancement help wanted won't fix
schwobr picture schwobr·11 Sep 2020·6 Comments

mlflow training loss not reported until end of run

bug / fix help wanted
david-waterworth picture david-waterworth·8 Sep 2020·8 Comments

ValueError: All dicts must have the same number of keys on model evaluation output.

DP bug / fix help wanted
Vichoko picture Vichoko·3 Sep 2020·3 Comments

CPU training is broken for more than one process

Working as intended bug / fix help wanted
tbenst picture tbenst·2 Sep 2020·10 Comments
👍1

How to free up the CUDA memory

question won't fix
mcemilg picture mcemilg·31 Aug 2020·6 Comments

Number of training steps

question won't fix
fperazzi picture fperazzi·28 Aug 2020·3 Comments
 Previous issues
Next issues