from datasets import load_dataset
from datasets import interleave_datasets

# Load as streaming = True
en_dataset = load_dataset('oscar', "unshuffled_deduplicated_en", split='train', streaming=True)
fr_dataset = load_dataset('oscar', "unshuffled_deduplicated_fr", split='train', streaming=True)

# Interleave
multilingual_dataset = interleave_datasets([en_dataset, fr_dataset])

# Shuffle
shuffled_dataset = multilingual_dataset.shuffle(seed=42, buffer_size=10_000)