Datasets:

openpecha
/

OCR-Google_Books

dataset_info:
  features:
    - name: filename
      dtype: string
    - name: label
      dtype: string
    - name: url
      dtype: string
    - name: BDRC_work_id
      dtype: string
    - name: char_len
      dtype: int64
    - name: script
      dtype: string
    - name: print_method
      dtype: string
  splits:
    - name: train
      num_bytes: 210467728
      num_examples: 601152
    - name: eval
      num_bytes: 26280512
      num_examples: 75136
    - name: test
      num_bytes: 26308535
      num_examples: 75168
  download_size: 76386563
  dataset_size: 263056775
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: eval
        path: data/eval-*
      - split: test
        path: data/test-*

Dataset Structure

Features: filename, label, url, BDORC_work_id, char_len, script, print_method
Splits: train, eval, test

📊 Split-wise Metadata

Split	# Samples	Total Chars (`char_len` sum)
Train	601,152	37,334,253
Eval	75,136	4,657,320
Test	75,168	4,666,128
Total	751,456	46,657,701

🏷️ Column Value Counts

print_method

Split	PrintMethod_Relief_WoodBlock	PrintMethod_Modern
Train	21,314	579,838
Eval	2,624	72,512
Test	2,565	72,603
Total	724,953	26,503

script

split	ScriptTibt	ScriptDbuCan	ScriptHani
Train	555,594	39,733	4,188
Eval	69,420	4,981	536
Test	69,343	5,093	546
Total	69,420	4,9807	5,270

🚀 Usage

from datasets import load_dataset
ds = load_dataset("openpecha/OCR-Google_Books", split="train")