masterful 0.3.6

Masterful AutoML Platform.

Masterful AutoML Platform

Masterful is an AutoML platform for the training of deep learning computer vision models.

It improves your model's accuracy - without the need for more labels - through robust implementations of semi-supervised learning, comprehensive regularization (including data augmentation, weight decay, and dropout control), and drop-in architectural enhancements.

Unlike many competing frameworks, Masterful's techniques are built to work with Masterful's purpose-built metalearner, relieving the developer of manual guessing and checking. This approach solves many of the issues found in black box optimization such as Bayesian optimization, Reinforcement Learning, or randomly sampled grid search.

Masterful also applies multiple techniques to minimize wall-clock time and GPU hours, including pushing nearly every data augmentation technique to the GPU using Masterful's purpose built augmentations; Masterful's custom training loop and scaffold model concept, and metalearning of optimal batch size, learning rate schedule, and epochs for your hardware.

The platform supports most types of classification, detection, and segmentation.

Currently available for Tensorflow2, with PyTorch support coming soon.