Exercise

Augmenting Monet

Perhaps you're still working on that machine learning model that identifies ocean scenes in paintings. You'd like to generate a few extra images to augment your existing data. After all, a human can tell that a painting is of an ocean even if the painting is upside-down: why shouldn't your machine learning model?

numpy is loaded as np, and the 3D Monet rgb_array is available.

Instructions 1/2

undefined XP

1
- Flip rgb_array so that it is the mirror image of the original, with the ocean on the right and grassy knoll on the left.

2
- Flip rgb_array so that it is upside down but otherwise remains the same.