README.md

metadata

language:
  - fa
size_categories:
  - 1K<n<10K
task_categories:
  - token-classification
dataset_info:
  features:
    - name: id
      dtype: int64
    - name: text
      dtype: string
    - name: relation
      dtype: string
    - name: e1
      dtype: string
    - name: e2
      dtype: string
    - name: is_reversed
      dtype: bool
    - name: raw
      dtype: string
  splits:
    - name: train
      num_bytes: 3306893
      num_examples: 8000
    - name: test
      num_bytes: 1113759
      num_examples: 2717
  download_size: 2331411
  dataset_size: 4420652
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*

Link to article: https://arxiv.org/abs/2005.06588

Downloaded data from farsbase.net