metadata
dataset_info:
features:
- name: dialog
sequence: string
- name: act
sequence: int64
- name: emotion
sequence: int64
- name: translated_dialogs
sequence: string
- name: input_ids
sequence: int64
- name: attention_mask
sequence: int64
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 285752784
num_examples: 11118
- name: validation
num_bytes: 25730415
num_examples: 1000
download_size: 12903418
dataset_size: 311483199
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*