Buffered Python generators for data augmentation

When using deep convolutional neural networks (CNNs) for image classification tasks, it’s common to apply several transformations to the images in order to augment the data and reduce overfitting. For example, images are often randomly cropped, mirrored, … Read more

Similar

Magnum Engine integrates Python and C++

During the past four months, Magnum began its adventure into the Python world. Not just with some autogenerated bindings and not just with some autogenerated Sphinx docs — that simply wouldn’t be Magnum enough. Brace yourselves, this article will show you... (more…)

Read more »