Viewing: diffsharp.github.io

[diffsharp-logo-text.png]

DiffSharp.Model Namespace

Contains types and functionality related to describing models.

Type/Module	Description
BatchNorm1d	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Applies Batch Normalization over a 2D or 3D input (a mini-batch of 1D inputs with optional additional channel dimension)
BatchNorm2d	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Applies Batch Normalization over a 4D input (a mini-batch of 2D inputs with optional additional channel dimension)
BatchNorm3d	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Applies Batch Normalization over a 5D input (a mini-batch of 3D inputs with optional additional channel dimension)
Conv1d	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] A model that applies a 1D convolution over an input signal composed of several input planes
Conv2d	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] A model that applies a 2D convolution over an input signal composed of several input planes
Conv3d	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] A model that applies a 3D convolution over an input signal composed of several input planes
ConvTranspose1d	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] A model that applies a 1D transposed convolution operator over an input image composed of several input planes.
ConvTranspose2d	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] A model that applies a 2D transposed convolution operator over an input image composed of several input planes.
ConvTranspose3d	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] A model that applies a 3D transposed convolution operator over an input image composed of several input planes.
Dropout	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] A model which during training, randomly zeroes some of the elements of the input tensor with probability p using samples from a Bernoulli distribution.
Dropout2d	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] A model which during training, randomly zero out entire channels. Each channel will be zeroed out independently on every forward call with probability p using samples from a Bernoulli distribution.
Dropout3d	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] A model which during training, randomly zero out entire channels. Each channel will be zeroed out independently on every forward call with probability p using samples from a Bernoulli distribution.
Linear	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] A model that applies a linear transformation to the incoming data: \(y = xA^T + b\)
LSTM	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Long short-term memory (LSTM) recurrent neural network.
LSTMCell	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Unit cell of a long short-term memory (LSTM) recurrent neural network. Prefer using the RNN class instead, which can combine RNNCells in multiple layers.
Mode	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Indicates the training or evaluation mode for a model.
Model	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png]
"> Model	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Represents a model, primarily a collection of named parameters and sub-models and a function governed by them.
ModelBase	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Represents the base class of all models.
Parameter	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Represents a parameter.
ParameterDict	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Represents a collection of named parameters.
RecurrentShape	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png]
RNN	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Recurrent neural network.
RNNCell	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Unit cell of a recurrent neural network. Prefer using the RNN class instead, which can combine RNNCells in multiple layers.
Sequential	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png]
VAE	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Variational auto-encoder
VAEBase	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Variational auto-encoder base
VAEMLP	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Variational auto-encoder with multilayer perceptron (MLP) encoder and decoder.
Weight	[copy-md.png] [copy-md-hover.png] [copy-xml.png] [copy-xml-hover.png] [github.png] [github-hover.png] Contains functionality related to generating initial parameter weights for models.