search

Lightning-AI / pytorch-lightning

Pretrain, finetune ANY AI model of ANY size on multiple GPUs, TPUs with zero code changes.
https://lightning.ai
Apache License 2.0
28.26k stars 3.38k forks source link

typeError unexpected closure #4940

Closed yikuanli closed 3 years ago

yikuanli commented 3 years ago

❓ Questions and Help

I am using pytorch 1.6 and pytorch lightning 1.0.8.

Before I upgrade my pytorch-lighing from 1.0.2 to 1.0.8, everything works fine, today I upgrade to 1.0.8 to try some new metrics, but got this error. 

main()

File "main.py", line 68, in main trainer.fit(model, trainloader, evalloader) File "/home/yikuan/anaconda/envs/py3/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 444, in fit results = self.accelerator_backend.train() File "/home/yikuan/anaconda/envs/py3/lib/python3.7/site-packages/pytorch_lightning/accelerators/gpu_accelerator.py", line 63, in train results = self.train_or_test() File "/home/yikuan/anaconda/envs/py3/lib/python3.7/site-packages/pytorch_lightning/accelerators/accelerator.py", line 74, in train_or_test results = self.trainer.train() File "/home/yikuan/anaconda/envs/py3/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 493, in train self.train_loop.run_training_epoch() File "/home/yikuan/anaconda/envs/py3/lib/python3.7/site-packages/pytorch_lightning/trainer/training_loop.py", line 561, in run_training_epoch batch_output = self.run_training_batch(batch, batch_idx, dataloader_idx) File "/home/yikuan/anaconda/envs/py3/lib/python3.7/site-packages/pytorch_lightning/trainer/training_loop.py", line 728, in run_training_batch self.optimizer_step(optimizer, opt_idx, batch_idx, train_step_and_backward_closure) File "/home/yikuan/anaconda/envs/py3/lib/python3.7/site-packages/pytorch_lightning/trainer/training_loop.py", line 470, in optimizer_step optimizer, batch_idx, opt_idx, train_step_and_backward_closure File "/home/yikuan/anaconda/envs/py3/lib/python3.7/site-packages/pytorch_lightning/accelerators/accelerator.py", line 130, in optimizer_step using_lbfgs=is_lbfgs File "/home/yikuan/anaconda/envs/py3/lib/python3.7/site-packages/pytorch_lightning/core/lightning.py", line 1270, in optimizer_step optimizer.step(closure=optimizer_closure) File "/home/yikuan/anaconda/envs/py3/lib/python3.7/site-packages/torch/optim/lr_scheduler.py", line 67, in wrapper return wrapped(*args, *kwargs) File "/home/yikuan/anaconda/envs/py3/lib/python3.7/site-packages/torch/autograd/grad_mode.py", line 15, in decorate_context return func(args, **kwargs) TypeError: step() got an unexpected keyword argument 'closure' (py3) yikuan@deepmedicine:~/project/version_control/HiBEHRT-BYOL$ TypeError: step() got an unexpected keyword argument 'closure'

github-actions[bot] commented 3 years ago

Hi! thanks for your contribution!, great first issue!

ananyahjha93 commented 3 years ago

Which optimizer and scheduler are you using? If you are using a custom optimizer, you need to update the code of the step() function to take in a closure. PyTorch optimizer class implements the step method with the closure parameter in the latest version.

yikuanli commented 3 years ago

thanks, I now upgrade it to the latest version and seems the latest bolts solve the problem.

ananyahjha93 commented 3 years ago

Oh, if you were using the LARS wrapper form bolts, then the latest version has the step method taking in a closure. That should solve it.

Should I close this issue then?

yikuanli commented 3 years ago

yes, thanks a lot