metadata
dataset_info:
features:
- name: commit_hash
dtype: string
- name: commit_message
dtype: string
- name: changed_files
sequence: string
- name: pre_change_chunks
dtype: string
- name: post_change_chunks
dtype: string
- name: relevant_chunks
sequence: string
splits:
- name: train
num_bytes: 12750624
num_examples: 56
download_size: 4435018
dataset_size: 12750624
configs:
- config_name: default
data_files:
- split: train
path: data/train-*