metadata

dataset_info:
  features:
    - name: text
      dtype: string
    - name: spans
      list:
        - name: end
          dtype: int64
        - name: label
          dtype: string
        - name: start
          dtype: int64
    - name: meta
      struct:
        - name: file_id
          dtype: string
        - name: sentence_id
          dtype: int64
  splits:
    - name: train
      num_bytes: 12633387
      num_examples: 61603
    - name: eval
      num_bytes: 1582593
      num_examples: 7700
    - name: test
      num_bytes: 1586745
      num_examples: 7701
    - name: overlapping
      num_bytes: 2465230
      num_examples: 8427
  download_size: 7023581
  dataset_size: 18267955
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: eval
        path: data/eval-*
      - split: test
        path: data/test-*
      - split: overlapping
        path: data/overlapping-*

SpaCy Grammar Error Detection Dataset - Binary Labels

This repository contains the SpaCy-formatted dataset for grammar error detection with binary labels.

Files:

train.json: Training data
eval.json: Evaluation data
test.json: Test data
overlapping.json: Sentences with overlapping error spans

Format:

Each file contains a JSON array of examples in SpaCy format, with text and span annotations.

Usage:

These files can be used directly with SpaCy's span categorization training.