search

AlibabaResearch / AdvancedLiterateMachinery

A collection of original, innovative ideas and algorithms towards Advanced Literate Machinery. This project is maintained by the OCR Team in the Language Technology Lab, Tongyi Lab, Alibaba Group.
Apache License 2.0
1.4k stars 169 forks source link

Can't pickle local object 'linear_scheduler.<locals>.lr_lambda' #45

Open WuChannn opened 1 year ago

WuChannn commented 1 year ago

when i use multiple gpus to train GeoLayoutLM, error as follows:

Can't pickle local object 'linear_scheduler..lr_lambda'

@wdp-007 @congyao @alibaba-oss Looking forward to your reply.

hossein-pdftron commented 1 year ago

Any update on this issue?

kenneys-bot commented 1 year ago

I had the same problem today when I deployed the project to train on multiple GPUs： LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0,1,2] File "/public/home/lab70432/workspace/env_sources/geolayoutlm/train.py", line 64, in <module> main() File "/public/home/lab70432/workspace/env_sources/geolayoutlm/train.py", line 60, in main trainer.fit(pl_module, datamodule=data_module) AttributeError: Can't pickle local object 'linear_scheduler.<locals>.lr_lambda' I don't have any solution to this problem yet.

Any update on this issue?

Have you solved the problem yet? Thanks.