---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: labels_str
    dtype: string
  - name: labels
    dtype: int64
  splits:
  - name: train
    num_bytes: 697282
    num_examples: 5928
  - name: eval
    num_bytes: 172941
    num_examples: 1482
  - name: test
    num_bytes: 249286
    num_examples: 2000
  download_size: 685486
  dataset_size: 1119509
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: eval
    path: data/eval-*
  - split: test
    path: data/test-*
---

This is a Hugging Face Dataset wrapper of the Sentipolc Twitter dataset (Basile et al., 2014)