`push_to_hub` a dataset dict with subsets and splits (e.g., GLUE)

You can now use push_to_hub to push multiple subsets of your dataset ! e.g.

dataset_subset1.push_to_hub("username/dataset_name", "subset1")
dataset_subset2.push_to_hub("username/dataset_name", "subset2")

# later

dataset_subset1 = load_dataset("username/dataset_name", "subset1")
dataset_subset2 = load_dataset("username/dataset_name", "subset2")

Each subset can be a DatasetDict made of multiple splits, or you can upload one split at a time:

dataset_subset1_train.push_to_hub("username/dataset_name", "subset1", split="train")
dataset_subset1_test.push_to_hub("username/dataset_name", "subset1", split="test")

# later

dataset_subset1_train = load_dataset("username/dataset_name", "subset1", split="train")
dataset_subset1_test = load_dataset("username/dataset_name", "subset1", split="test")
4 Likes