Skip to content

issue with --jit_mode_eval enabled in trainer commandline #19973

@sywangyi

Description

@sywangyi

System Info

  • transformers version: 4.24.0.dev0
  • Platform: Linux-5.15.0-46-generic-x86_64-with-glibc2.31
  • Python version: 3.9.13
  • Huggingface_hub version: 0.10.1
  • PyTorch version (GPU?): 1.12.1+cu102 (True)
  • Tensorflow version (GPU?): not installed (NA)
  • Flax version (CPU?/GPU?/TPU?): not installed (NA)
  • Jax version: not installed
  • JaxLib version: not installed
  • Using GPU in script?:
  • Using distributed or parallel set-up in script?:

Who can help?

Trainer: @sgugger

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

python3 examples/pytorch/text-classification/run_glue.py --model_name_or_path /skyrex01/wangyi/output/mrpc/ --task_name mrpc --do_eval --max_seq_length 128 --output_dir /skyrex01/wangyi/output/mrpc/inference1/ --overwrite_output_dir True --fp16 --jit_mode_eval

Expected behavior

jit success and run the jit traced model.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions