| dataset_info: | |
| features: | |
| - name: image | |
| dtype: image | |
| - name: ground_truth | |
| dtype: string | |
| - name: labels | |
| dtype: string | |
| splits: | |
| - name: train | |
| num_bytes: 339139900 | |
| num_examples: 4000 | |
| download_size: 332704349 | |
| dataset_size: 339139900 | |
| configs: | |
| - config_name: default | |
| data_files: | |
| - split: train | |
| path: data/train-* | |
| deepseek ocr labeled |