/code2seq-1

Re-implementation of "CODE2SEQ: GENERATING SEQUENCES FROM STRUCTURED REPRESENTATIONS OF CODE"

Primary LanguagePythonMIT LicenseMIT

code2seq [WIP]

A PyTorch re-implementation code for "code2seq: Generating Sequences from Structured Representations of Code"

Requirements

Please see requirements.txt

Usage

  • notebooks/preparation.ipynb for downloading dataset, making some directories etc.

  • notebooks/code2seq.ipynb for training and evaluating the model.

Memo

  • Beam search is not implemented.
  • GCP AI Platform Notebooks is used to train model.
  • AI Platform Notebooks requires google_compute_engine api so please install this before installing other packages if you use AI Platform Notebooks.