Skip to content

Latest commit

 

History

History
19 lines (18 loc) · 900 Bytes

README.md

File metadata and controls

19 lines (18 loc) · 900 Bytes
Raw

The NLC2CMD competition@NeurIPS 2020

Competition Link

Training models

  1. Install libraries: pytorch, fairseq
    pip install -r requirements.txt
  2. Download NL2Bash data and AInix data in train/data directory
  3. Setup submodule TellinaTool ./0.setup_submodule.sh
  4. Convert the file format of AInix
    ./1.convert_ainix_data.sh (make a sh file for doing 2, 3, and 4 at once)
  5. Run preprocess
    ./2.preprocess.sh
  6. Train template predictor
    ./3.train_template_predictor.sh
  7. Train argument predictor
    ./4.train_argument_predictor.sh