vis4d.pl.callbacks.callback_wrapper¶
Wrapper to connect PyTorch Lightning callbacks.
Functions
|
Get loss_module from pl module. |
|
Get model from pl module. |
|
Wrap pl.Trainer and pl.LightningModule into Trainer. |
Classes
|
Wrapper to connect vis4d callbacks to pytorch lightning callbacks. |
- class CallbackWrapper(callback)[source]¶
Wrapper to connect vis4d callbacks to pytorch lightning callbacks.
Init class.
- on_test_batch_end(trainer, pl_module, outputs, batch, batch_idx, dataloader_idx=0)[source]¶
Wait for on_test_batch_end PL hook to call ‘process’.
- Return type:
None
- on_test_epoch_end(trainer, pl_module)[source]¶
Wait for on_test_epoch_end PL hook to call ‘evaluate’.
- Return type:
None
- on_test_epoch_start(trainer, pl_module)[source]¶
Hook to run at the start of a testing epoch.
- Return type:
None
- on_train_batch_end(trainer, pl_module, outputs, batch, batch_idx)[source]¶
Hook to run at the end of a training batch.
- Return type:
None
- on_train_batch_start(trainer, pl_module, batch, batch_idx)[source]¶
Called when the train batch begins.
- Return type:
None
- on_train_epoch_end(trainer, pl_module)[source]¶
Hook to run at the end of a training epoch.
- Return type:
None
- on_train_epoch_start(trainer, pl_module)[source]¶
Hook to run at the start of a training epoch.
- Return type:
None
- on_validation_batch_end(trainer, pl_module, outputs, batch, batch_idx, dataloader_idx=0)[source]¶
Wait for on_validation_batch_end PL hook to call ‘process’.
- Return type:
None
- on_validation_epoch_end(trainer, pl_module)[source]¶
Wait for on_validation_epoch_end PL hook to call ‘evaluate’.
- Return type:
None