| dataset_info: | |
| features: | |
| - name: scenes | |
| sequence: | |
| sequence: float32 | |
| - name: language_annotation | |
| dtype: string | |
| splits: | |
| - name: train | |
| num_bytes: 41820172461 | |
| num_examples: 7966221 | |
| download_size: 4545543118 | |
| dataset_size: 41820172461 | |
| configs: | |
| - config_name: default | |
| data_files: | |
| - split: train | |
| path: data/train-* | |