Training doesn't start

[Psoriaz]

I'm trying to train a model on a custom dataset. The training begins, but stops at the moment of generating the dataset. The last lines that are printed to the console:

Resolving data files: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2117/2117 [00:00<00:00, 42802.73it/s] Resolving data files: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 605/605 [00:00<00:00, 46075.35it/s] Resolving data files: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 304/304 [00:00<00:00, 55909.34it/s] Downloading data: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2119/2119 [00:00<00:00, 38616.95files/s] Downloading data: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 607/607 [00:00<00:00, 45081.68files/s] Downloading data: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 306/306 [00:00<00:00, 52478.11files/s] Generating train split: 2116 examples [00:00, 9723.71 examples/s] Generating validation split: 604 examples [00:00, 17335.75 examples/s] Generating test split: 303 examples [00:00, 16679.23 examples/s]

What should I do? How to understand where exactly the problem is?

[Rakshith12-pixel]

any updates???? im stuck with the same issue

[Psoriaz]

The problem has been resolved.

Steps taken:

1. Created a virtual environment
2. Cloned the repository
3. I went to the donut directory and wrote pip install .
4. Downgraded the version of transformers to 4.18.0, timm to 0.5.4, tokenizers to 0.12.1, torch to 2.0.0+cu117

Now everything works fine!

[Rakshith12-pixel]

ERROR: Could not find a version that satisfies the requirement torch==2.0.0+cu117 (from versions: 1.11.0, 1.12.0, 1.12.1, 1.13.0, 1.13.1, 2.0.0, 2.0.1, 2.1.0, 2.1.1, 2.1.2, 2.2.0, 2.2.1, 2.2.2, 2.3.0, 2.3.1, 2.4.0, 2.4.1, 2.5.0, 2.5.1)
which one shuld be used?
THanks

[Rakshith12-pixel]

Anyways, i used the versions you mentioned, kept torch at Version: 2.3.0 and then had to downgrade protobuf to 3.20.x and the training has started. Thanks @Psoriaz