Skip to content

Latest commit

 

History

History
40 lines (25 loc) · 707 Bytes

File metadata and controls

40 lines (25 loc) · 707 Bytes

Data

Write-up: https://dudeperf3ct.github.io/projects/train_llm_part0/

Pre-requisites

  • uv

Getting Started

  1. Install the dependencies:
uv sync
  1. Configure the settings in config.yaml as needed.

  2. Run the dataset parsing script from root of the repo. It downloads the dataset on the first run if not already present at data/raw_dataset/<name-of-dataset> path.

cd ..
python -m codellm_data.main parse
  1. Run the dataset downloading script from root of the repo.
cd ..
python -m codellm_data.main download

Architecture

Parser

parser

Downloader

downloader