metadata
dataset_info:
features:
- name: tweet_text
dtype: string
- name: disaster_type
dtype: int64
splits:
- name: train
num_bytes: 16972808
num_examples: 107359
download_size: 10414888
dataset_size: 16972808
configs:
- config_name: default
data_files:
- split: train
path: data/train-*