bitorch.models¶

This submodule contains a number of adapted model architectures that use binary / quantized weights and activations.

To define a new model, use the Model base class as a super class.

Functions

`model_from_name`	Return a model by the given name.
`model_names`	Get the list of model names.
`register_custom_model`	Register a custom (external) model in bitorch.

Modules

`bitorch.models.base`
`bitorch.models.common_layers`
`bitorch.models.densenet`
`bitorch.models.dlrm`
`bitorch.models.lenet`
`bitorch.models.meliusnet`
`bitorch.models.model_hub`
`bitorch.models.quicknet`
`bitorch.models.resnet`
`bitorch.models.resnet_e`	Resnet_E implementation from "Back to Simplicity: How to Train Accurate BNNs from Scratch?" paper.

bitorch