Stateful Predict

Predict responses using a trained recurrent neural network

expand all in page
  • Stateful Predict block

Libraries:
Deep Learning Toolbox / Deep Neural Networks

Description

The Stateful Predict block predicts responses for the data at the input by using the trained recurrent neural network specified through the block parameter. This block allows loading of a pretrained network into the Simulink® model from a MAT file or from a MATLAB® function. This block updates the state of the network with every prediction.

To reset the state of recurrent neural network to its initial state, place the Stateful Predict block inside a Resettable Subsystem (Simulink) block and use the Reset control signal as trigger.

Examples

Predict and Update Network State in Simulink

Predict and Update Network State in Simulink

Predict responses for a trained recurrent neural network in Simulink® by using the Stateful Predict block. This example uses a pretrained long short-term memory (LSTM) network.

Limitations

  • CPU acceleration using the Intel® MKL-DNN library and GPU acceleration using the NVIDIA® CuDNN or TensorRT libraries are not supported for Stateful Predict blocks that use a dlnetwork object.

Ports

Input

expand all

The input ports of the Stateful Predict block takes the names of the input layers of the network loaded. Based on the network loaded, the input to the predict block can be sequence or time series data.

The dimensions of the numeric arrays containing the sequences depend on the type of data.

InputDescription
Vector sequencess-by-c matrices, where s is the sequence length, and c is the number of features of the sequences.
2-D image sequencesh-by-w-by-c-by-s arrays, where h, w, and c correspond to the height, width, and number of channels of the images, respectively, and s is the sequence length.

Output

expand all

The outputs port of the Stateful Predict block takes the names of the output layers of the network loaded. Based on the network loaded, the output of the Stateful Predict block can represent predicted scores or responses.

For sequence-to-label classification, the output is a K-by-N matrix, where K is the number of classes, and N is the number of observations.

For sequence-to-sequence classification problems, the output is a K-by-S matrix of scores, where K is the number of classes, and S is the total number of time steps in the corresponding input sequence.

Parameters

expand all

Specify the source for the trained recurrent neural network. The trained network must have at least one recurrent layer (for example, an LSTM network). Select one of the following:

  • Network from MAT-file— Import a trained recurrent neural network from a MAT file containing a dlnetwork object.

  • Network from MATLAB function— Import a pretrained recurrent neural network from a MATLAB function.

Programmatic Use

Block Parameter: Network
Type: character vector, string
Values: 'Network from MAT-file' | 'Network from MATLAB function'
Default: 'Network from MAT-file'

This parameter specifies the name of the MAT file that contains the trained recurrent neural network to load. If the file is not on the MATLAB path, use the Browse button to locate the file.

Dependencies

To enable this parameter, set the Network parameter to Network from MAT file.

Programmatic Use

Block Parameter: NetworkFilePath
Type: character vector, string
Values: MAT file path or name
Default: 'untitled.mat'

This parameter specifies the name of the MATLAB function for the pretrained recurrent neural network.

Dependencies

To enable this parameter, set the Network parameter to Network from MATLAB function.

Programmatic Use

Block Parameter: NetworkFunction
Type: character vector, string
Values: MATLAB function name
Default: 'untitled'

Since R2026a

  • Auto mode –– This option automatically selects the simulation mode based on the model configuration and code generation compatibility. It uses Code generation whenever code generation is supported and switches to Interpreted execution when the model is incompatible for code generation.

  • Interpreted execution –– Simulate model using the MATLAB interpreter. This option shortens startup time but has a slower simulation speed than Code generation. In this mode, you can debug the source code of the block.

  • Code generation –– Simulate model using generated code.

    The first time you simulate a model that contains a Stateful Predict block, the software generates binary code or C/C++ MATLAB executable (MEX) code from the block and integrates this code with the model. The Stateful Predict block uses the same infrastructure as MATLAB Coder™, which you use to generate C/C++ code from MATLAB code outside of Simulink. The generated code is reused for subsequent simulations, as long as the model does not change. This option requires additional startup time, but the speed of the subsequent simulations is comparable to Interpreted execution.

If your Simulink model is set up to simulate in Normal mode and if you use Stateful Predict block via code generation, individual Stateful Predict blocks can take advantage of acceleration via code generation. If the Simulink model is set up to simulate in accelerator mode or any of the other target modes, Stateful Predict block's Simulate using parameter does not have any effect. However, there is one exception, if the model is set up to simulate in accelerator mode and you use Stateful Predict block with interpreted execution mode, the block will run in interpreted execution mode.

Programmatic Use

Block Parameter: SimulateUsing
Type: character vector, string
Values: "Auto mode" | "Interpreted execution" | "Code generation"
Default: "Auto mode"

The Sample time parameter specifies when the block computes a new output value during simulation. For details, see Specify Sample Time (Simulink).

Specify the Sample time parameter as a scalar when you do not want the output to have a time offset. To add a time offset to the output, specify the Sample time parameter as a 1-by-2 vector where the first element is the sampling period and the second element is the offset.

By default, the Sample time parameter value is -1 to inherit the value.

Programmatic Use

Block Parameter: SampleTime
Type: character vector
Values: scalar | vector
Default: '-1'

This parameter specifies the input data format expected by the trained dlnetwork.

Data format, specified as a string scalar or a character vector. Each character in the string must be one of these dimension labels:

  • "S" — Spatial

  • "C" — Channel

  • "B" — Batch

  • "T" — Time

  • "U" — Unspecified

For example, for an array containing a batch of sequences where the first, second, and third dimension correspond to channels, observations, and time steps, respectively, you can specify that it has the format "CBT".

You can specify multiple dimensions labeled "S" or "U". You can use the labels "C", "B", and "T" once each, at most. The software ignores singleton trailing "U" dimensions after the second dimension.

For more information, see Deep Learning Data Formats.

By default, the parameter uses the data format that the network expects.

Dependencies

To enable this parameter, set the Network parameter to Network from MAT-file to import a trained dlnetwork object from a MAT file.

Programmatic Use

Block Parameter: InputDataFormats
Type: character vector, string
Values: For a network with one or more inputs, use character vector in the form of: {'inputlayerName1', 'SSC'; 'inputlayerName2', 'SSCB'; ...}'. For a network with no input layer and multiple input ports, use character vector in the form of: '{'inputportName1/inport1, 'SSC'; 'inputportName2/inport2, 'SSCB'; ...}'.
Default: Data format that the network expects. For more information, see Deep Learning Data Formats.

Extended Capabilities

expand all

Version History

Introduced in R2021a

expand all

See Also

| |