To load the dataset on your local device:
# Local loading
from datasets import load_dataset
dataset = load_dataset("/home/think3/Desktop/1. MSCOCO_captions_dataset_edited/dataset_test_jsonl/ImageCaptions.py", split='train[{}:]'.format(1),cache_dir="dataset_test_jsonl/caching")
To load the dataset from Huggingface:
# Test the remote repo:
from datasets import load_dataset
dataset = load_dataset("Arabic-Clip/mscoco_jsonl_full", split='train[:]', cache_dir="cache/remote")