`validation_epoch_end` problem, how can I fix it?

[OrilinZ]

/usr/lib/python3/dist-packages/requests/init.py:89: RequestsDependencyWarning: urllib3 (2.0.3) or chardet (3.0.4) doesn't match a supported version!
warnings.warn("urllib3 ({}) or chardet ({}) doesn't match a supported "
2023-06-15 11:20:10.232 | INFO | diffusion_categorical:init:105 - [compute transition matrix]: gaussian
2023-06-15 11:20:25.947 | INFO | diffusion_categorical:init:125 - [trainsition matrix]: torch.Size([1000, 256, 256])
2023-06-15 11:20:25.947 | INFO | diffusion_categorical:init:128 - [Construct transition matrices for q(x_t|x_start)]
2023-06-15 11:20:26.259 | INFO | diffusion_categorical:init:140 - [tilde(Q)t]: torch.Size([1000, 256, 256])
CometLogger will be initialized in offline mode
GPU available: True (cuda), used: True
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs
Traceback (most recent call last):
File "main.py", line 256, in
trainer.fit(d3pm, ckpt_path=args.resume)
File "/home/zhz/.local/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 531, in fit
call._call_and_handle_interrupt(
File "/home/zhz/.local/lib/python3.8/site-packages/pytorch_lightning/trainer/call.py", line 41, in _call_and_handle_interrupt
return trainer.strategy.launcher.launch(trainer_fn, *args, trainer=trainer, **kwargs)
File "/home/zhz/.local/lib/python3.8/site-packages/pytorch_lightning/strategies/launchers/subprocess_script.py", line 91, in launch
return function(*args, **kwargs)
File "/home/zhz/.local/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 570, in _fit_impl
self._run(model, ckpt_path=ckpt_path)
File "/home/zhz/.local/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 923, in _run
_verify_loop_configurations(self)
File "/home/zhz/.local/lib/python3.8/site-packages/pytorch_lightning/trainer/configuration_validator.py", line 36, in _verify_loop_configurations
__verify_train_val_loop_configuration(trainer, model)
File "/home/zhz/.local/lib/python3.8/site-packages/pytorch_lightning/trainer/configuration_validator.py", line 84, in __verify_train_val_loop_configuration
raise NotImplementedError(
NotImplementedError: Support for validation_epoch_end has been removed in v2.0.0. DiffusionModel implements this method. You can use the on_validation_epoch_end hook instead. To access outputs, save them in-memory as instance attributes. You can find migration examples in Lightning-AI/pytorch-lightning#16520.

[ljh0v0]

Hi,

Seems Pytorch-lightning removed the validation _epoch_end method in version 2.0.0. You can either downgrade the Pytorch-lightning version to 1.8.1 or adapt the validation_epoch_end part referring to this.

Good luck!

Jiahe

[liuyi-193]

You can use the "on_validation_epoch_end" hook instead.