Package gmlp implements a model composed by basic MLP layers with gating mechanism.

Package mlpmixer implements the MLP-Mixer (Tolstikhin et al., 2021).

Package sgu implements the Spatial Gating Unit (SGU).

Apply fn recursively to every sub-models as well as self.

Buf creates a new Buffer Node.

Dump saves a serialized object to a stream.

DumpToFile saves a serialized object to a file.

ForEachParam iterate all the parameters of a model also exploring the sub-parameters recursively.

ForEachParamStrict iterate all the parameters of a model without exploring the sub-models.

Load uses Gob to deserialize objects to memory.

LoadFromFile uses Gob to deserialize objects files to memory.

NewParam returns a new param.

ZeroGrad set the gradients of all model's parameters (including sub-params) to zeros.

Buffer is a type of Node that do not require gradients but that can be serialized similarly to any other parameters.

Module must be embedded into all neural models.

Model is implemented by all neural network architectures.

ParamsTraverser allows you to define a custom procedure to traverse the parameters of a model.

StandardModel consists of a model that implements a Forward variadic function that accepts mat.Tensor and returns a slice of mat.Tensor.