| dataset_info: | |
| features: | |
| - name: sequence | |
| dtype: string | |
| - name: label | |
| dtype: int64 | |
| splits: | |
| - name: train | |
| num_bytes: 2893542 | |
| num_examples: 3424 | |
| - name: valid | |
| num_bytes: 740946 | |
| num_examples: 857 | |
| - name: test | |
| num_bytes: 2594652 | |
| num_examples: 2529 | |
| download_size: 2914341 | |
| dataset_size: 6229140 | |
| configs: | |
| - config_name: default | |
| data_files: | |
| - split: train | |
| path: data/train-* | |
| - split: valid | |
| path: data/valid-* | |
| - split: test | |
| path: data/test-* | |
| From this [paper](https://www.sciencedirect.com/science/article/pii/S2405805X25001905). 1 is high expression, 0 is low. |