metadata
dataset_info:
features:
- name: tweet
dtype: string
- name: sentiment
dtype: string
- name: dialect
dtype: string
- name: label
dtype:
class_label:
names:
'0': Non sarcastic
'1': sarcastic
splits:
- name: train
num_bytes: 2500183
num_examples: 12548
- name: test
num_bytes: 614876
num_examples: 3000
download_size: 1749568
dataset_size: 3115059
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*