| --- |
| dataset_info: |
| features: |
| - name: image |
| dtype: image |
| - name: bboxes |
| sequence: |
| sequence: float64 |
| - name: labels |
| sequence: int64 |
| - name: words |
| sequence: string |
| - name: split |
| dtype: string |
| splits: |
| - name: train |
| num_bytes: 32034973965.125 |
| num_examples: 80863 |
| download_size: 0 |
| dataset_size: 32034973965.125 |
| configs: |
| - config_name: default |
| data_files: |
| - split: train |
| path: data/train-* |
| --- |
| # Dataset Card for "doclaynet_processed" |
| |
| Clean version of [DocLayNet](https://github.com/DS4SD/DocLayNet) ready for finetuning. |
