metadata
dataset_info:
features:
- name: message
dtype: string
- name: message_type
dtype: string
- name: message_id
dtype: int64
- name: conversation_id
dtype: int64
- name: cluster
dtype: float64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 10058978
num_examples: 8904
download_size: 2338890
dataset_size: 10058978
configs:
- config_name: default
data_files:
- split: train
path: data/train-*