focalLossLayer

Create focal loss layer using focal loss function

expand all in page

Description

A focal loss layer predicts object classes using focal loss.Add the focal loss layer to train an object detection, semantic segmentation, or a classification network when imbalance exists between foreground and background classes. To compensate for class imbalance, the focal loss function multiplies the cross entropy function with a modulating factor that increases the sensitivity of the network to misclassified observations.

Creation

Syntax

layer = focalLossLayer

layer = focalLossLayer(Name,Value)

Description

example

layer= focalLossLayercreates a focal loss layer for deep learning networks. For information on how to use focal loss layer in an object detection network, seeCreate SSD Object Detection Network.

example

layer= focalLossLayer(Name,Value)sets properties of the focal loss layer by using one or more name-value pair arguments. Enclose each property name in quotes.

For example,focalLossLayer('Name','focalloss')creates a focal loss layer with the name'focalloss'and the specified balancing and focusing parameters.

Properties

expand all

`Alpha`—Balancing parameter
0.25(default) |positive real number

Balancing parameter of the focal loss function, specified as a positive real number. TheAlphavalue scales the loss function linearly and is typically set to0.25. If you decreaseAlpha, increaseGamma.

`Gamma`—Focusing parameter
2.0(default) |positive real number

Focusing parameter of the focal loss function, specified as a positive real number. Increasing the value ofGammaincreases the sensitivity of the network to misclassified observations.

`Classes`—Object classes to detect
`'auto'`(default) |字符串向量|categorical vector|cell array of character vectors

Object classes that the network is trained to detect, specified as a string vector, categorical vector, cell array of character vectors, or'auto'. When you setClassesto'auto', the classes are automatically set at training time. When you specify a string vector or cell array of character vectors, then the elements ofClassesare sorted according to the output of thecategoriesfunction.

Data Types:string|categorical|cell|char

`Name`—Layer name
`''`(default) |特征向量|string scalar

Layer name, specified as a character vector or a string scalar. ForLayerarray input, thetrainNetwork,assembleNetwork,layerGraph, anddlnetworkfunctions automatically assign names to layers with name''.

Data Types:char|string

Examples

collapse all

Create Focal Loss Layer

Open Live Script

Specify class names.

classes = ["Vehicle","Background"];

Specify the balancing paramete, and focusing parameter of the focal loss function. Create a focal loss layer named "focallosslayer" for the two classes, displaying results.

layer = focalLossLayer('Classes',classes,'Name','focallosslayer')

layer = FocalLossLayer with properties: Name: 'focallosslayer' Hyperparameters Gamma: 2 Alpha: 0.2500 Classes: [2x1 categorical] LossFunction: 'focalLoss'

Create 2-D Semantic Segmentation Network with Focal Loss Layer

This example uses:

Open Live Script

Create a DeepLab v3+ network based on ResNet-18.

imageSize = [480 640 3]; numClasses = 5; network ='resnet18'; lgraph = deeplabv3plusLayers(imageSize,numClasses,network,'DownsamplingFactor',16)

lgraph = LayerGraph with properties: Layers: [100x1 nnet.cnn.layer.Layer] Connections: [113x2 table] InputNames: {'data'} OutputNames: {'classification'}

Display the output layer of the network. The output layer of the DeepLab v3+ network is aPixelClassificationLayerthat uses cross-entropy loss to predict the categorical label for every pixel in an input 2-D image.

lgraph.Layers(end)

ans = PixelClassificationLayer with properties: Name: 'classification' Classes: 'auto' ClassWeights: 'none' OutputSize: 'auto' Hyperparameters LossFunction: 'crossentropyex'

Replace the outputPixelClassificationLayerwith theFocalLossLayerto handle imbalanced classes in data.

layer = focalLossLayer("Name","focalloss"); lgraph = replaceLayer(lgraph,"classification",layer);

Display the network.

analyzeNetwork(lgraph);

Create 3-D Semantic Segmentation Network with Focal Loss Layer

This example uses:

Open Live Script

Create a 3-D U-Net network for semantic segmentation by usingunet3dLayersfunction. Set the encoder-decoder depth to 2 and specify the number of output channels for the first convolution layer as 16.

imageSize = [128 128 128 3]; numClasses = 5; lgraph = unet3dLayers(imageSize,numClasses,'EncoderDepth',2,...'NumFirstEncoderFilters',16); figure plot(lgraph)

Figure contains an axes object. The axes object contains an object of type graphplot.

Create a focal loss layer and replace theSegmentation-Layerin the network with the focal loss layer. The layer predicts the categorical label for every voxel in an input 3-D volume.

layer = focalLossLayer("Name","focalloss"); lgraph = replaceLayer(lgraph,"Segmentation-Layer",layer)

lgraph = LayerGraph with properties: Layers: [40x1 nnet.cnn.layer.Layer] Connections: [41x2 table] InputNames: {'ImageInputLayer'} OutputNames: {'focalloss'}

Display the network.

analyzeNetwork(lgraph);

More About

expand all

Focal Loss

The focal loss function is based on cross-entropy loss. Focal loss compensates for class imbalance by using a modulating factor that emphasizes hard negatives during training.

The focal loss function,L, used by thefocalLossLayer对象之间的损失imageYand the corresponding ground truthTis given by:

$L = - \frac{1}{M} \sum_{m = 1}^{M} \sum_{k = 1}^{K} T_{m k} α {(1 - Y_{m k})}^{γ} \ln (Y_{m k})$

Kis the number of classes.Mis the number of observations, which is typically the number of elements along the first two dimensions ofYmultiplied by the number of anchor boxes. The function used by thefocalLossLayerobject ensures that the input to the natural logarithm is greater than 0.

References

[1] Lin, Tsung-Yi, Priya Goyal, Ross Girshick, Kaiming He, and Piotr Dollar. "Focal Loss for Dense Object Detection." In 2017IEEE^®International Conference on Computer Vision (ICCV),2999–3007. Venice: IEEE, 2017. https://doi.org/10.1109/ICCV.2017.324.

Extended Capabilities

GPU Code Generation
Generate CUDA® code for NVIDIA® GPUs using GPU Coder™.

To generate CUDA^®or C++ code by using GPU Coder™, you must first construct and train a deep neural network. Once the network is trained and evaluated, you can configure the code generator to generate code and deploy the convolutional neural network on platforms that use NVIDIA^®or ARM^®GPU processors. For more information, seeDeep Learning with GPU Coder(GPU Coder).

For this layer, you can generate code that takes advantage of the NVIDIA CUDA deep neural network library (cuDNN), NVIDIA TensorRT™ high performance inference library, or the ARMCompute Libraryfor Mali GPU.

Version History

Introduced in R2020a

focalLossLayer

Description

Creation

Syntax

Description

Properties

`Alpha`—Balancing parameter
0.25(default) |positive real number

`Gamma`—Focusing parameter
2.0(default) |positive real number

`Classes`—Object classes to detect
`'auto'`(default) |字符串向量|categorical vector|cell array of character vectors

`Name`—Layer name
`''`(default) |特征向量|string scalar

Examples

Create Focal Loss Layer

Create 2-D Semantic Segmentation Network with Focal Loss Layer

Create 3-D Semantic Segmentation Network with Focal Loss Layer

More About

Focal Loss

References

Extended Capabilities

GPU Code Generation
Generate CUDA® code for NVIDIA® GPUs using GPU Coder™.

Version History

See Also

Topics

focalLossLayer

Description

Creation

Syntax

Description

Properties

Alpha—Balancing parameter0.25(default) |positive real number

Gamma—Focusing parameter2.0(default) |positive real number

Classes—Object classes to detect'auto'(default) |字符串向量|categorical vector|cell array of character vectors

Name—Layer name''(default) |特征向量|string scalar

Examples

Create Focal Loss Layer

Create 2-D Semantic Segmentation Network with Focal Loss Layer

Create 3-D Semantic Segmentation Network with Focal Loss Layer

More About

Focal Loss

References

Extended Capabilities

GPU Code GenerationGenerate CUDA® code for NVIDIA® GPUs using GPU Coder™.

Version History

See Also

Topics

`Alpha`—Balancing parameter
0.25(default) |positive real number

`Gamma`—Focusing parameter
2.0(default) |positive real number

`Classes`—Object classes to detect
`'auto'`(default) |字符串向量|categorical vector|cell array of character vectors

`Name`—Layer name
`''`(default) |特征向量|string scalar

GPU Code Generation
Generate CUDA® code for NVIDIA® GPUs using GPU Coder™.