invertransforms 0.1.1

A library which turns torchvision transformations invertible and replayable.

invertransforms

A library which turns torchvision transformations invertible and replayable.

Installation

pip install invertransforms

Usage

Simply replace previous torchvision import statements and enjoy the addons.

# from torchvision import transforms as T
import invertransforms as T

transform = T.Compose([
  T.RandomCrop(size=256),
  T.ToTensor(),
])

img_tensor = transform(img)

# invert
img_again = transform.invert(img_tensor)

# replay
img_tensor2 = transform.replay(img2)

All transformations have an inverse transformation attached to it.

inv_transform = transform.inverse()
img_inv = inv_transform(img)

Notes:

If a transformation is random, it is necessary to apply it once before calling invert or inverse(). Otherwise it will raise InvertibleError. On the otherhand, replay can be called before, it will simply set the randomness on its first call.

Use Case

This library can be particularly useful in following situations:

• Reverting your model outputs in order to stack predictions
• Applying the same transformations on two different images
• blah-blah

Features

• Invert any transformations even random ones
• Replay any transformations even random ones
• All classes extends its torchvision transformation equivalent class
• Extensive unit testing
• blah-blah

Documentation

The above features are explained in more detail in invertransforms' documentation.