search

Jun-jie-Huang / ExeDS

Source code and data for our paper at DaSH@EMNLP-22: "Execution-based Evaluation for Data Science Code Generation Models"
MIT License
7 stars 1 forks source link

The checkpoint weights #1

Closed IssamLaradji closed 1 year ago

IssamLaradji commented 1 year ago

do you know where I can find the checkpoint weights for the trained PyMT5 and JupyT5? Thanks.

Jun-jie-Huang commented 1 year ago

Hi, the PyMT5 and JupyT5 model is provided by the authors in the paper of PyMT5: multi-mode translation of natural language and Python code with transformers and Training and Evaluating a Jupyter Notebook Data Science Assistant. I'm not sure whether they released the checkpoint weights, but you can contact the authors of the above two papers to obtain the checkpoint weights.

IssamLaradji commented 1 year ago

Thanks!