awacke1's picture
Update app.py
05df1af
raw
history blame
573 Bytes
from datasets import load_dataset
raw_datasets = load_dataset("allocine")
raw_datasets.save_to_disk("my-arrow-datasets")
from datasets import load_from_disk
arrow_datasets_reloaded = load_from_disk("my-arrow-datasets")
arrow_datasets_reloaded
for split, dataset in raw_datasets.items():
dataset.to_csv(f"my-dataset-{split}.csv", index=None)
data_files = {
"train": "my-dataset-train.csv",
"validation": "my-dataset-validation.csv",
"test": "my-dataset-test.csv",
}
csv_datasets_reloaded = load_dataset("csv", data_files=data_files)
csv_datasets_reloaded