To load the dataset on your local device:

```py
# 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:

```py
# Test the remote repo:

from datasets import load_dataset

dataset = load_dataset("Arabic-Clip/mscoco_jsonl_full", split='train[:]', cache_dir="cache/remote")
```