Pickle error and OOM when upgrading to 1.2.0
See original GitHub issueWhen upgrading from 1.1.6 to 1.2.0, I notices 2 changes
- Significant increase of gpu memory
- Pickle error in my module class (the object
metric_fn
is sure not pickable but the same code worked fine in 1.1.6)
Do you have an idea what changes in 1.2.0 may cause the issues ? Any suggestion for the memory problem ?
My pseudo code
class ClfModule(pl.LightningModule):
def __init__(self, model, tokenizer):
self.model = model
self.tokenizer = tokenizer
self.metric_fn = not_pickable_object()
I am using Huggingface transformers and datasets
Issue Analytics
- State:
- Created 3 years ago
- Comments:7 (3 by maintainers)
Top Results From Across the Web
Pickle error and OOM when upgrading to 1.2.0 #6173 - GitHub
When upgrading from 1.1.6 to 1.2.0, I notices 2 changes. Significant increase of gpu memory; Pickle error in my module class (the object ......
Read more >Pickle dump huge file without memory error - Stack Overflow
I was having the same issue. I use joblib and work was done. In case if someone wants to know other possibilities. save...
Read more >pickle.load can't get attribute python - You.com | The Search ...
I had the same error using pandas 1.3.2 in the server while 1.2 in my client. Downgrading pandas to 1.2 solved the problem....
Read more >Deployment Platform: Render - Part 1 (2019) - Fast.ai forums
Hi everyone, I'm the founder of Render, a new cloud provider that makes it trivial to deploy webapps in production.
Read more >Changelog — PyTorch Lightning 1.8.5 documentation
Fixed a pickling error when using RichProgressBar together with checkpointing ... Reset the dataloaders on OOM failure in batch size finder to use...
Read more >
Top Related Medium Post
No results found
Top Related StackOverflow Question
No results found
Troubleshoot Live Code
Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start Free
Top Related Reddit Thread
No results found
Top Related Hackernoon Post
No results found
Top Related Tweet
No results found
Top Related Dev.to Post
No results found
Top Related Hashnode Post
No results found
@phihung My OOM went away with upgrade to v1.2.1 and using DDP. Please try.
Dear @angadkalra,
Were using using AMP ? In 1.2.0, there were a bug and AMP wasn’t applied properly. It is solved in 1.2.1. Which could explain your OOM in 1.2.0 and not in 1.2.1.
Best, T.C