MKYucel/zero_shot_corruption_benchmarks

[Elsevier Image and Vision Computing] How robust are discriminatively trained zero-shot learning models?

Python