metadata

license: mit
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
dataset_info:
  features:
    - name: method
      dtype: string
    - name: clean_method
      dtype: string
    - name: doc
      dtype: string
    - name: comment
      dtype: string
    - name: method_name
      dtype: string
    - name: extra
      struct:
        - name: repo_name
          dtype: string
        - name: path
          dtype: string
        - name: license
          dtype: string
        - name: size
          dtype: int64
    - name: imports
      sequence: string
    - name: imports_info
      dtype: string
    - name: cluster_imports_info
      dtype: string
    - name: libraries
      sequence: string
    - name: libraries_info
      dtype: string
    - name: id
      dtype: int64
  splits:
    - name: train
      num_bytes: 5373034269
      num_examples: 2916582
  download_size: 2492962682
  dataset_size: 5373034269
tags:
  - code-generation
pretty_name: 'CodeGen4Libs '
size_categories:
  - 1M<n<10M

Dataset Card for FudanSELab CodeGen4Libs Code Retrieval Library

Dataset Description

Repository: GitHub Repository
Paper: CodeGen4Libs: A Two-stage Approach for Library-oriented Code Generation

Dataset Summary

This dataset is the code retrieval library used in the ASE2023 paper titled "CodeGen4Libs: A Two-stage Approach for Library-oriented Code Generation".

Additional Information

Citation Information

@inproceedings{ase2023codegen4libs,
  author       = {Mingwei Liu and Tianyong Yang and Yiling Lou and Xueying Du and Ying Wang and and Xin Peng},
  title        = {{CodeGen4Libs}: A Two-stage Approach for Library-oriented Code Generation},
  booktitle    = {38th {IEEE/ACM} International Conference on Automated Software Engineering,
                  {ASE} 2023, Kirchberg, Luxembourg, September 11-15, 2023},
  pages        = {0--0},
  publisher    = {{IEEE}},
  year         = {2023},
}