metadata
dataset_info:
features:
- name: caller
dtype: string
- name: text
dtype: string
- name: act_tag
dtype:
class_label:
names:
'0': B
'1': D
'2': F
'3': Q
'4': S
- name: conversation_no
dtype: string
- name: speaker_change
dtype:
class_label:
names:
'0': no_change
'1': change
splits:
- name: train
num_bytes: 5519384
num_examples: 75067
- name: validation
num_bytes: 1223647
num_examples: 16433
- name: test
num_bytes: 1202512
num_examples: 16702
download_size: 2521941
dataset_size: 7945543
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*