metadata
dataset_info:
features:
- name: repo_name
dtype: string
- name: pr_number
dtype: int64
- name: pr_title
dtype: string
- name: pr_description
dtype: string
- name: author
dtype: string
- name: date_created
dtype: timestamp[ns, tz=UTC]
- name: date_merged
dtype: timestamp[ns, tz=UTC]
- name: previous_commit
dtype: string
- name: pr_commit
dtype: string
- name: query
dtype: string
- name: filepath
dtype: string
- name: before_content
dtype: string
- name: after_content
dtype: string
- name: label
dtype: int64
splits:
- name: train
num_bytes: 117579951237
num_examples: 2074433
download_size: 58125931847
dataset_size: 117579951237
configs:
- config_name: default
data_files:
- split: train
path: data/train-*