DAMO-ConvAI/star/pretrain

STAR

This is the project containing source code for the paper [STAR: SQL Guided Pre-Training for Context-dependent Text-to-SQL Parsing]

Create conda environment

The following commands.

Create conda environment star:

• In our experiments, we use torch==1.7.0 with CUDA version 11.0

• We use four GeForce A-100 for our pre-trained experiments.

  conda create -n star python=3.6
  conda activate star
  pip install torch==1.7.0+cu110 -f https://download.pytorch.org/whl/torch_stable.html
  pip install -r requirements.txt
  

Unzip pretraining dataset

Download and move the pretaining data file pretrain_data.txt into the directory datasets.

Training

(It may takes two days on four Tesla V100-PCIE-32GB GPU.)

    python pretain_inbatch.py

Saving STAR model

    python save_model.py

Then you can get the trained model and its configuration (at least containing model.bin and config.json) under pretrained/sss direction.