README.md

metadata

dataset_info:
  features:
    - name: code
      dtype: string
    - name: path
      dtype: string
    - name: quality_prob
      dtype: float64
    - name: learning_prob
      dtype: float64
    - name: filename
      dtype: string
    - name: kind
      dtype: string
  splits:
    - name: train
      num_bytes: 29415279411
      num_examples: 3450448
  download_size: 9929877988
  dataset_size: 29415279411
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Dataset Card for "clean_code_data"

High quality code data filtering from the stack and pypi. The stack data is pulled from starcoder data, so it has some filtering applied already.

All data was cleaned to remove code licenses and other headers, and filtered for quality and learning value.