plotWeights

Class: Autoencoder

(To be removed) Plot a visualization of the weights for the encoder of an autoencoder

plotWeights will be removed in a future release. For more information, see Transition Legacy Neural Network Code to dlnetwork Workflows.

For advice on updating your code, see Version History.

Syntax

plotWeights(autoenc)

h = plotWeights(autoenc)

Description

plotWeights(autoenc) visualizes the weights for the autoencoder, autoenc.

h = plotWeights(autoenc) returns a function handle h, for the visualization of the encoder weights for the autoencoder, autoenc.

Input Arguments

autoenc — Trained autoencoder

Autoencoder object

Trained autoencoder, specified as an Autoencoder object.

Output Arguments

h — Image object

handle

Image object, returned as a handle.

Examples

Visualize Learned Features

Load the training data.

X = digitTrainCellArrayData;

The training data is a 1-by-5000 cell array, where each cell contains a 28-by-28 matrix representing a synthetic image of a handwritten digit.

Train an autoencoder with a hidden layer of 25 neurons.

hiddenSize = 25;
autoenc = trainAutoencoder(X,hiddenSize, ...
  'L2WeightRegularization',0.004, ...
  'SparsityRegularization',4, ...
  'SparsityProportion',0.2);

Visualize the learned features.

plotWeights(autoenc);

Tips

• plotWeights allows the visualization of the features that the autoencoder learns. Use it when the autoencoder is trained on image data. The visualization of the weights has the same dimensions as the images used for training.

Version History

Introduced in R2015b

R2026a: plotWeights will be removed

plotWeights will be removed in a future release.

To learn more about autoencoder workflows, see these examples and topics:

• Build Networks with Deep Network Designer

• Time Series Anomaly Detection Using Deep Learning

• Detect Anomalies Using Wavelet Scattering with Autoencoders

• Train Variational Autoencoder (VAE) to Generate Images

• List of Deep Learning Layer Blocks and Subsystems

• Code Generation for Deep Learning Networks

For more information, see Transition Legacy Neural Network Code to dlnetwork Workflows.