Grouped Convolution 2D Layer

R2026b

Grouped 2-D convolutional layer

Since R2026b

expand all in page
  • Grouped Convolution 2D Layer block

Libraries:
Deep Learning Toolbox / Deep Learning Layers / Convolution and Fully Connected Layers

Description

The Grouped Convolution 2D Layer separates the input channels into groups and applies sliding convolutional filters. Use grouped convolutional layers for channel-wise separable (also known as depth-wise separable) convolution.

This block accepts 2-D image data in the SSC format (three dimensions corresponding to two spatial dimensions and one channel dimension, in that order). It convolves the input by moving the filters along the input vertically and horizontally and computing the dot product of the weights and the input, and then adding a bias term. The block combines the convolutions for each group independently. If the number of groups is equal to the number of channels, then this block performs channel-wise convolution.

The exportNetworkToSimulink function generates this block to represent a groupedConvolution2dLayer object.

Limitations

  • The Layer parameter does not support groupedConvolution2dLayer objects that have the PaddingValue property set to "symmetric-exclude-edge".

  • The Layer parameter does not support groupedConvolution2dLayer objects that have the DilationFactor property set to a value other than 1.

Ports

Input

expand all

Input data on which to perform the convolution operation. The data must have three dimensions or fewer and be larger than the convolutional filter size.

If the data has fewer than three dimensions, the block treats the input as having trailing singleton dimensions. For example, for h-by-w input, the block treats the input as an h-by-w-by-1 (SSC) array.

Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | fixed point

Output

expand all

The result of convolving the input data over the first two (spatial) dimensions for each group of channels. The output data has three dimensions corresponding to two spatial dimensions and one channel dimension, in that order.

Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | fixed point

Parameters

expand all

To edit block parameters interactively, use the Property Inspector. From the Simulink® Toolstrip, on the Simulation tab, in the Prepare gallery, select Property Inspector.

Main

Specify the name of a workspace variable that contains a groupedConvolution2dLayer object from a trained network. The Grouped Convolution 2D Layer block configures itself by using the properties of the object and calculates the block output by using the learnable parameters of the object.

Programmatic Use

Block Parameter: Layer
Type: workspace variable
Values: groupedConvolution2dLayer object
Default: 'layerObject'

Data format for the input data. The options use the same notation as the fmt argument of the dlarray object, except layer blocks do not support the Batch (B) dimension and instead assume an observation number of 1.

Programmatic Use

Block Parameter: DataFormat
Type: character vector
Values: 'SSC'
Default: 'SSC'

Data Types

Minimum value of the output range that the software checks.

The software uses the minimum value to perform:

Tips

Output minimum does not saturate or clip the actual output signal. Use the Saturation (Simulink) block instead.

Programmatic Use

To set the block parameter value programmatically, use the set_param (Simulink) function.

Parameter: OutMin
Values: '[]' (default) | scalar in quotes

Maximum value of the output range that the software checks.

The software uses the maximum value to perform:

Tips

Output maximum does not saturate or clip the actual output signal. Use the Saturation (Simulink) block instead.

Programmatic Use

To set the block parameter value programmatically, use the set_param (Simulink) function.

Parameter: OutMax
Values: '[]' (default) | scalar in quotes

Choose the data type for the output. The type can be inherited, specified directly, or expressed as a data type object such as Simulink.NumericType. When you select Inherit: Inherit via internal rule, the software chooses a data type to balance numerical accuracy, performance, and generated code size, while taking into account the properties of the embedded target hardware.

Programmatic Use

To set the block parameter value programmatically, use the set_param (Simulink) function.

Parameter: OutDataTypeStr
Values: 'Inherit: Inherit via internal rule' (default) | 'Inherit: Keep MSB' | 'Inherit: Keep LSB' | 'Inherit: Inherit via back propagation' | 'Inherit: Same as first input' | 'Inherit: Same as accumulator' | <data type expression>

Select this parameter to prevent Fixed-Point Tool from overriding the Output data type you specify on the block. For more information, see Use Lock Output Data Type Setting (Fixed-Point Designer).

Programmatic Use

To set the block parameter value programmatically, use the set_param (Simulink) function.

Parameter: LockScale
Values: 'off' (default) | 'on'

Specify the rounding mode for the output value of the block, generated by the final fixed-point cast operation. For the rounding mode for the other internal block operations, see Internal integer rounding mode. For more information, see Rounding Modes (Fixed-Point Designer).

Block parameters always round to the nearest representable value. To control the rounding of a block parameter, enter an expression using a MATLAB® rounding function in the mask field.

Programmatic Use

To set the block parameter value programmatically, use the set_param (Simulink) function.

Parameter: RndMeth
Values: 'Floor' (default) | 'Ceiling' | 'Convergent' | 'Nearest' | 'Round' | 'Simplest' | 'Zero'

Specify whether integer overflows saturate or wrap for the output value of the block, which is generated by the final fixed-point cast operation. For the method of overflow action of the internal block operations, see Internal saturate on integer overflow.

  • on — Overflows saturate to either the minimum or maximum value that the data type can represent.

  • off — Overflows wrap according to the rules of the data type.

For example, the maximum value that the signed 8-bit integer int8 can represent is 127. Any block operation result greater than the maximum value causes overflow of the 8-bit integer.

  • With this parameter selected, the block output saturates at 127. Similarly, the block output saturates at a minimum output value of –128.

  • With this parameter cleared, the software interprets the overflow-causing value as int8, which can produce an unintended result. For example, a block result of 130 (binary 1000 0010) expressed as int8 is –126.

Tips

  • Set this parameter to on when your model has a possible overflow and you want explicit saturation protection in the generated code.

  • To optimize the efficiency of your generated code, keep the default off setting for this parameter. Using the default setting also helps you avoid overspecifying how the block handles out-of-range signals. For more information, see Troubleshoot Signal Range Errors (Simulink).

  • In general, the code generation process can detect when overflow is not possible. In this case, the code generator does not produce saturation code.

Programmatic Use

To set the block parameter value programmatically, use the set_param (Simulink) function.

Parameter: SaturateOnIntegerOverflow
Values: 'off' (default) | 'on'

The block casts the value of the Weights property of the object that you specify with the Layer parameter to this data type. The type can be inherited, specified directly, or expressed as a data type object such as Simulink.NumericType.

Programmatic Use

To set the block parameter value programmatically, use the set_param (Simulink) function.

Parameter: WeightsDataTypeStr
Values: 'Inherit: Inherit via back propagation' (default) | 'Inherit: Inherit from 'Constant value'' | <data type expression>

The block casts the value of the Bias property of the object that you specify with the Layer parameter to this data type. The type can be inherited, specified directly, or expressed as a data type object such as Simulink.NumericType.

Programmatic Use

To set the block parameter value programmatically, use the set_param (Simulink) function.

Parameter: BiasDataTypeStr
Values: 'Inherit: Inherit via back propagation' (default) | 'Inherit: Inherit from 'Constant value'' | <data type expression>

Choose the data type of the accumulator for each Convolution block inside the Grouped Convolution 2D Layer block. The type can be inherited, specified directly, or expressed as a data type object such as Simulink.NumericType. When you select Inherit: Inherit via internal rule, the software chooses a data type to balance numerical accuracy, performance, and generated code size, while taking into account the properties of the embedded target hardware.

Programmatic Use

To set the block parameter value programmatically, use the set_param (Simulink) function.

Parameter: AccumDataTypeStr
Values: 'Inherit: Inherit via internal rule' (default) | 'Inherit: Same as first input' | <data type expression>

Specify the rounding mode for internal fixed-point computations, except for the output operation. For the rounding mode of the output operation, see Output integer rounding mode. For more information, see Rounding Modes (Fixed-Point Designer).

Block parameters always round to the nearest representable value. To control the rounding of a block parameter, enter an expression using a MATLAB rounding function in the mask field.

Programmatic Use

To set the block parameter value programmatically, use the set_param (Simulink) function.

Parameter: InternalRndMeth
Values: 'Floor' (default) | 'Ceiling' | 'Convergent' | 'Nearest' | 'Round' | 'Simplest' | 'Zero'

Specify whether integer overflows saturate or wrap for the internal operations of the block. For the method of overflow action of the block output, see Output saturate on integer overflow.

  • on — Overflows saturate to either the minimum or maximum value that the data type can represent.

  • off — Overflows wrap according to the rules of the data type.

For example, the maximum value that the signed 8-bit integer int8 can represent is 127. Any block operation result greater than the maximum value causes overflow of the 8-bit integer.

  • With this parameter selected, the block output saturates at 127. Similarly, the block output saturates at a minimum output value of –128.

  • With this parameter cleared, the software interprets the overflow-causing value as int8, which can produce an unintended result. For example, a block result of 130 (binary 1000 0010) expressed as int8 is –126.

Tips

  • Set this parameter to on when your model has a possible overflow and you want explicit saturation protection in the generated code.

  • To optimize the efficiency of your generated code, keep the default off setting for this parameter. Using the default setting also helps you avoid overspecifying how the block handles out-of-range signals. For more information, see Troubleshoot Signal Range Errors (Simulink).

  • In general, the code generation process can detect when overflow is not possible. In this case, the code generator does not produce saturation code.

Programmatic Use

To set the block parameter value programmatically, use the set_param (Simulink) function.

Parameter: InternalSaturateOnIntegerOverflow
Values: 'off' (default) | 'on'

Execution

Specify the discrete interval between sample time hits or specify another type of sample time, such as continuous (0) or inherited (-1). For more options, see Types of Sample Time (Simulink).

By default, the block inherits its sample time based on the context of the block within the model.

Programmatic Use

To set the block parameter value programmatically, use the set_param (Simulink) function.

Parameter: SampleTime
Data Types: char
Values: '-1' | scalar
Default: '-1'

Programmatic Use: Add Block

To add the block programmatically, use the add_block function with one of these library block paths as the source.

"dlconvfclib/Grouped Convolution 2D Layer"

Example: add_block("dlconvfclib/Grouped Convolution 2D Layer","myModel/myBlock",MakeNameUnique="on")

For information about programmatic access to individual block parameters, see the Programmatic Use section of parameter descriptions on this page.

Tips

  • To improve performance of generated code for a Grouped Convolution 2D Layer block, select the Automatically schedule for-loops (Embedded Coder) model configuration parameter.

Extended Capabilities

expand all

C/C++ Code Generation
Generate C and C++ code using Simulink® Coder™.

Version History

Introduced in R2026b

See Also

| |

Topics