---
dataset_info:
  features:
  - name: page_content
    dtype: string
  - name: metadata
    struct:
    - name: page
      dtype: int64
    - name: source
      dtype: string
  splits:
  - name: train
    num_bytes: 36605083
    num_examples: 16869
  download_size: 22765390
  dataset_size: 36605083
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---


language:
  - Turkish
pretty_name: Text data from fictions

This dataset created from 30+ Turkish classic fictions