Download dataset_info.json from taewonkoo/stack_cube_v2_rlds: direct link, hf CLI and curl.
- Browser
- Download file 544 Bytes
-
https://huggingface.co/datasets/taewonkoo/stack_cube_v2_rlds/resolve/main/dataset_info.json
- Command line
-
hf download hf://datasets/taewonkoo/stack_cube_v2_rlds/dataset_info.json
-
curl -L -o dataset_info.json https://huggingface.co/datasets/taewonkoo/stack_cube_v2_rlds/resolve/main/dataset_info.json
544 Bytes
| { | |
| "citation": "Dataset converted with Forge", | |
| "description": "Dataset converted with Forge", | |
| "fileFormat": "tfrecord", | |
| "moduleName": "forge_dataset.forge_dataset", | |
| "name": "forge_dataset", | |
| "releaseNotes": { | |
| "1.0.0": "Converted with Forge" | |
| }, | |
| "splits": [ | |
| { | |
| "filepathTemplate": "{DATASET}-{SPLIT}.{FILEFORMAT}-{SHARD_X_OF_Y}", | |
| "name": "train", | |
| "numBytes": "1196938542", | |
| "shardLengths": [ | |
| "10", | |
| "10", | |
| "10", | |
| "10", | |
| "10" | |
| ] | |
| } | |
| ], | |
| "version": "1.0.0" | |
| } |