While converting gpt-2-simple model, its unable to save 'pytorch/pytorch_model.bin'

[anandmg101]

Hi, trying to convert gpt2-simple fine tuned model. I'm getting this error. I guess its not able to save the pytorch model. Not sure why.

Any suggestions?

Save PyTorch model to pytorch/pytorch_model.bin Traceback (most recent call last): File "/home/XXXX/XXXX/djgpt2env/bin/transformers-cli", line 8, in sys.exit(main()) File "/home/XXXX/XXXX/djgpt2env/lib/python3.8/site-packages/transformers/commands/transformers_cli.py", line 51, in main service.run() File "/home/XXXX/XXXX/djgpt2env/lib/python3.8/site-packages/transformers/commands/convert.py", line 152, in run convert_gpt2_checkpoint_to_pytorch(self._tf_checkpoint, self._config, self._pytorch_dump_output) File "/home/XXXX/XXXX/djgpt2env/lib/python3.8/site-packages/transformers/models/gpt2/convert_gpt2_original_tf_checkpoint_to_pytorch.py", line 45, in convert_gpt2_checkpoint_to_pytorch torch.save(model.state_dict(), pytorch_weights_dump_path) File "/home/XXXX/XXXX/djgpt2env/lib/python3.8/site-packages/torch/serialization.py", line 369, in save with _open_file_like(f, 'wb') as opened_file: File "/home/XXXX/XXXX/djgpt2env/lib/python3.8/site-packages/torch/serialization.py", line 230, in _open_file_like return _open_file(name_or_buffer, mode) File "/home/XXXX/XXXX/djgpt2env/lib/python3.8/site-packages/torch/serialization.py", line 211, in init super(_open_file, self).init(open(name, mode)) FileNotFoundError: [Errno 2] No such file or directory: 'pytorch/pytorch_model.bin'

[anandmg101]

Found a work around, in case anyone is having the same issue.

from transformers import GPT2Model, GPT2Config config = GPT2Config.from_json_file('./checkpoint/run1/hparams.json') model = GPT2Model.from_pretrained('./checkpoint/run1/model-955.index', from_tf=True, config=config) model.save_pretrained('./pytorch', saved_model=True)

This saves the config and pytorch_model.bin files