gpt2-tensorflow-to-pytorch-converter

This repository contains a quick-to-use script to convert GPT-2 models from TensorFlow to PyTorch model format.

Usage

1. Collect all your TensorFlow model files into a singular directory, i.e. these files:

   model-<number>.meta
   vocab.bpe
   model-<number>.data-00000-of-00001
   model-<number>.index
   checkpoint
   counter
   encoder.json
   hparams.json
   

2. Clone the repo, install prerequisites with i.e. pip install -r requirements.txt if needed.

3. Run the script:

   python convert_model.py /path/to/your/model/files
   

4. The converted PyTorch model will be saved in the ./converted_model directory.

Notes

Have fun, I probably won't be updating this one much.

License

This project is licensed under the MIT License.

Contribute

All code improvements are welcome. This should at least work on all TF1.x-based GPT-2 architecture models.

About

• Flying from the mind of FlyingFathead
• Digital ghost code by ChaosWhisperer