---
language:
- cs
dataset_info:
  features:
  - name: tokens
    sequence: string
  - name: input_ids
    sequence: int32
  - name: labels
    dtype: int64
  - name: attention_mask
    sequence: int8
  splits:
  - name: train
    num_bytes: 5433513
    num_examples: 10416
  - name: validation
    num_bytes: 97082
    num_examples: 190
  - name: test
    num_bytes: 88690
    num_examples: 184
  download_size: 2505889
  dataset_size: 5619285
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---
This is a Czech Relationship extraction dataset downloaded from http://hdl.handle.net/11234/1-3265 and converted into HuggingFace dataset dictionary.