metadata

license: apache-2.0
arxiv: 2001.00059
pipeline_tag: fill-mask
tags:
  - code
  - cubert

CuBERT: Learning and Evaluating Contextual Embedding of Source Code

Overview

This model is the unofficial HuggingFace version of "CuBERT". In particular, this version comes from gs://cubert/20210711_Python/pre_trained_model_epochs_2__length_512. It was trained 2021-07-11 for 2 epochs with a 512 token context window on the Python BigQuery dataset. I manually converted the Tensorflow checkpoint to PyTorch, the tokenizer to a HuggingFace tokenizer, and have uploaded them here. All credit goes to Aditya Kanade, Petros Maniatis, Gogul Balakrishnan, and Kensen Shi.

Citation:

@inproceedings{cubert,
author    = {Aditya Kanade and
             Petros Maniatis and
             Gogul Balakrishnan and
             Kensen Shi},
title     = {Learning and evaluating contextual embedding of source code},
booktitle = {Proceedings of the 37th International Conference on Machine Learning,
               {ICML} 2020, 12-18 July 2020},
series    = {Proceedings of Machine Learning Research},
publisher = {{PMLR}},
year      = {2020},
}