composer.models.efficientnetb0#

Modules

The EfficientNet model family is a set of convolutional neural networks that can be used as the basis for a variety of vision tasks, but were initially designed for image classification. The model family was designed to reach the highest accuracy for a given computation budget during inference by simultaneously scaling model depth, model width, and image resolution according to an empirically determined scaling law.

See the Model Card for more details.

Classes

EfficientNetB0

A ComposerClassifier wrapper around the EfficientNet-b0 architecture.

Hparams

These classes are used with yahp for YAML-based configuration.