Sungha Choi comments

Results 5 comments of


                                            Sungha Choi

NonMatchingChecksumError for sun397

@M3Dade Hi, have you resolved the "wrong checksum" issue? Best,

FileNotFoundError: Couldn't find remote file with version master at https://raw.githubusercontent.com/huggingface/datasets/master/datasets/glue/glue.py. Please provide a valid version and a valid dataset name cp: cannot stat 'outputs/full_finetuning/all_results.json': No such file or directory

Hi @bf-yang Did you solve the problem you mentioned? I'm currently experiencing the same problem. Thank you!

[BUG] assert all_groups_norm > 0 | Error related to Bf16 optimizer it seems

I encountered the same issue. Is there any solution? > ... /deepspeed/runtime/bf16_optimizer.py", line 312, in step > [rank0]: assert all_groups_norm > 0. > [rank0]: AssertionError deepspeed 0.15.0 transformers 4.44.2

[BUG] assert all_groups_norm > 0 | Error related to Bf16 optimizer it seems

> I encountered the same issue. Is there any solution? > > > ... /deepspeed/runtime/bf16_optimizer.py", line 312, in step > > [rank0]: assert all_groups_norm > 0. > > [rank0]: AssertionError...

After sync batch norm is applied, more gpu memory is consumed

Hi @vacancy, Thanks a lot for your reply :) I have tested sync batch norm on deeplab-resnet based segmentation task. When I applied sync batch norm, it consumes about 30-40%...