when the training start, the process just stop at 2/66672 (steps),while the training process seems to continue, because after a while, the validation began, and the validation process bar still didn’t show up, the training bar didn’t move.
I test my code in colab, there is no problem when running in colab environment, I uninstall all packages to ensure all the package version is correct ,but It didn’t work. could anyone can help me
the matrix I use is seqeval,and I used my own dataset.
my code is alomost the same as the official notebook.
link:Google Colab
here is my environment info:
-
transformers
version:4.12.5 - Platform:pytorch
- Python version:3.8
- PyTorch version (GPU?):1.12.0-cuda11.3
- Tensorflow version (GPU?):
- Using GPU in script?:yes
- Using distributed or parallel set-up in script?:no