metadata
dataset_info:
features:
- name: commit_id
dtype: string
- name: commit_subject
dtype: string
- name: commit_message
dtype: string
- name: commit_query
dtype: string
- name: query_file_path
dtype: string
- name: true_documents
list: string
- name: true_document_paths
list: string
- name: secondary_documents
list: 'null'
- name: negative_documents
list: string
- name: hard_negative_documents
list: string
splits:
- name: train
num_bytes: 6121137
num_examples: 86
download_size: 483155
dataset_size: 6121137
configs:
- config_name: default
data_files:
- split: train
path: data/train-*