Permission Denied Error When Saving Checkpoint During Training

[Tanman2001]

System Info

• transformers version: 4.37.1
• Platform: Windows-10-10.0.22631-SP0
• Python version: 3.11.1
• Huggingface_hub version: 0.20.3
• Safetensors version: 0.4.2
• Accelerate version: 0.26.1
• PyTorch version (GPU?): 2.2.1+cpu (False)

Who can help?

No response

Information

• The official example scripts
• My own modified scripts

Tasks

• An officially supported task in the examples folder (such as GLUE/SQuAD, ...)
• My own task or dataset (give details below)

Reproduction

1. Follow exactly the Image Classification example in the HuggingFace docs: https://huggingface.co/docs/transformers/tasks/image_classification
2. Error thrown at trainer.train()
   

Expected behavior

Should train normally.

Issue and possible solution identified on forum earlier this year: https://discuss.huggingface.co/t/permission-issues-when-saving-model-checkpoint/70109/2

Removing (or commenting out, like pictured below) lines 2417-2420 of transformers/trainer.py resolves issue without further errors, but may have complications for other tasks.