example.py

from datasets import load_dataset

# Load the 300 text-image labelled pairs
ds = load_dataset("galatolo/TeTIm-Eval", "captioned")
print(ds)
for i, e in zip(range(0, 10), ds["test"]): print(f"{i}) {e}")

# Load the 2500 labelled pairs
ds = load_dataset("galatolo/TeTIm-Eval", "uncaptioned")
print(ds)
for i, e in zip(range(0, 10), ds["test"]): print(f"{i}) {e}")