| from datasets import load_dataset | |
| def main(): | |
| ds = load_dataset("Teklia/IAM-line") | |
| print(ds) | |
| sample = ds["train"][0] | |
| print("Keys:", sample.keys()) | |
| print("Text:", sample["text"]) | |
| # image is a PIL object inside HF datasets | |
| img = sample["image"] | |
| print("Image size:", img.size) | |
| if __name__ == "__main__": | |
| main() | |