Explain network predictions using Grad-CAM
returns the gradient-weighted class activation mapping (Grad-CAM) map of the change in the
classification score of image
scoreMap = gradCAM(
X, when the network
net evaluates the class score for the class given by
label. Use this function to explain network predictions and check
that your network is focusing on the right parts of an image.
The Grad-CAM interpretability technique uses the gradients of the classification score with respect to the final convolutional feature map. The parts of an image with a large value for the Grad-CAM map are those that most impact the network score for that class.
Use this syntax to compute the Grad-CAM map for image or pixel classification tasks.
returns the Grad-CAM importance map using a reduction function.
scoreMap = gradCAM(
reductionFcn is a function handle that reduces the output activations
of the reduction layer to a scalar value. This scalar fulfills the role of the class score
for classification tasks, and generalizes the Grad-CAM technique to nonclassification
tasks, such as regression.
gradCAM function computes the Grad-CAM map by differentiating
the reduced output of the reduction layer with respect to the features in the feature
gradCAM automatically selects reduction and feature layers to
use when computing the map. To specify these layers, use the
Use this syntax to compute the Grad-CAM map for nonclassification tasks.
Explore Network Classifications Using Grad-CAM
gradCAM to visualize which parts of an image are important to the classification decision of a network.
Import the pretrained network SqueezeNet.
net = squeezenet;
Import the image and resize it to match the input size for the network.
X = imread("laika_grass.jpg"); inputSize = net.Layers(1).InputSize(1:2); X = imresize(X,inputSize);
Display the image.
Classify the image to get the class label.
label = classify(net,X)
label = categorical toy poodle
gradCAM to determine which parts of the image are important to the classification result.
scoreMap = gradCAM(net,X,label);
Plot the result over the original image with transparency to see which areas of the image contribute most to the classification score.
figure imshow(X) hold on imagesc(scoreMap,'AlphaData',0.5) colormap jet
The network focuses predominantly on the back of the dog to make the classification decision.
Compute Grad-CAM Map for Image Regression Network
Use Grad-CAM to visualize which parts of an image are most important to the predictions of an image regression network.
Load the sample data, which consists of synthetic images of handwritten digits. The third output contains the corresponding angles of rotation of the digits, in degrees.
rng default [XTrain,~,YTrain] = digitTrain4DArrayData; [XTest,~,YTest] = digitTest4DArrayData; numTrainImages = numel(YTrain); idx = randperm(numTrainImages,20);
Construct an image regression network that can predict the rotation of an image.
layers = [ ... imageInputLayer([28 28 1],'Name','input') convolution2dLayer(12,25,'Name','conv') reluLayer('Name','relu') fullyConnectedLayer(1,'Name','fc') regressionLayer('Name','output')];
Specify the training options.
options = trainingOptions('sgdm', ... 'InitialLearnRate',0.001, ... 'Verbose',false, ... 'Plots','training-progress');
Train the network.
net = trainNetwork(XTrain,YTrain,layers,options);
Evaluate the performance of the network on a test image.
testDigit = XTest(:,:,:,idx(4));
predict to predict the angle of rotation and compare the predicted rotation to the true rotation.
predRotation = predict(net,testDigit)
predRotation = single -47.5497
trueRotation = YTest(idx(4))
trueRotation = -40
Visualize the regions of the image most important to the network prediction using
gradCAM. Select the ReLU layer as the feature layer and the fully connected layer as the reduction layer.
featureLayer = 'relu'; reductionLayer = 'fc';
Define the reduction function. The reduction function must reduce the output of the reduction layer to a scalar value. The Grad-CAM map displays the importance of different parts of the image to that scalar. In this regression problem, the network predicts the angle of rotation of the image. Therefore, the output of the fully connected layer is already a scalar value and so the reduction function is just the identity function.
reductionFcn = @(x)x;
Compute the Grad-CAM map.
scoreMap = gradCAM(net,testDigit,reductionFcn, ... 'ReductionLayer',reductionLayer, ... 'FeatureLayer',featureLayer);
Display the Grad-CAM map over the test image.
ax(1) = subplot(1,2,1); imshow(testDigit) title("True Rotation = " + trueRotation + '\newline Pred Rotation = ' + round(predRotation,0)) colormap(ax(1),'gray') ax(2) = subplot(1,2,2); imshow(testDigit) hold on imagesc(scoreMap) colormap(ax(2),'jet') title("GradCAM") hold off
The Grad-CAM map shows that the network is focusing on the area in the bottom left, which is where the tail of the digit would be if the image had zero rotation. The map suggests that to predict the negative rotation, the network is using the empty space.
net — Trained network
Trained network, specified as a
dlnetwork object. You can get a
trained network by importing a pretrained network or by training your own network using
trainNetwork function or custom training. For
more information about pretrained networks, see Pretrained Deep Neural Networks.
X — Input data
numeric array |
Input data, specified as a numeric array or formatted
X must have size equal to the input size of the network.
label — Class label
categorical | character vector | string scalar | numeric index
Class label to use for calculating the Grad-CAM map for image classification and semantic segmentation tasks, specified as a categorical, a character vector, a string scalar, a numeric index, or a vector of these values.
dlnetwork objects, you must specify
as a categorical or a numeric index.
If you specify
label as a vector, the software calculates the
feature importance for each class label independently. In that case,
scoreMap(:,:,k) corresponds to the map for
gradCAM function sums the spatial dimensions of the
reduction layer for class
label. Therefore, you can specify
label as the classes of interest for semantic segmentation tasks
gradCAM returns the Grad-CAM importance for each pixel.
reductionFcn — Reduction function
Reduction function, specified as a function handle. The reduction function reduces
the output activations of the reduction layer to a single value and must reduce a
dlarray object to a
dlarray scalar. This scalar
fulfills the role of
label in classification tasks, and generalizes
the Grad-CAM technique to nonclassification tasks, such as regression.
Grad-CAM uses the reduced output activations of the reduction layer to compute the gradients for the importance map.
Specify optional pairs of arguments as
the argument name and
Value is the corresponding value.
Name-value arguments must appear after other arguments, but the order of the
pairs does not matter.
Before R2021a, use commas to separate each name and value, and enclose
Name in quotes.
computes the Grad-CAM map with respect to layers
'prob', executes the calculation on the GPU, and upsamples the
resulting map to the same size as the input image using bicubic interpolation.
FeatureLayer — Name of feature layer
string | character vector
Name of the feature layer to extract the feature map from when computing the Grad-CAM map, specified as a string or character vector. For most tasks, use the last ReLU layer with nonsingleton spatial dimensions or the last layer that gathers the outputs of ReLU layers (such as depth concatenation or addition layers). If your network does not contain any ReLU layers, specify the name of the final convolutional layer that has nonsingleton spatial dimensions in the output.
The default value is the final layer with nonsingleton spatial dimensions. Use the
analyzeNetwork function to examine your network and select the
ReductionLayer — Name of reduction layer
string | character vector
Name of the reduction layer to extract output activations from when computing the
Grad-CAM map, specified as a string or character vector. For classification tasks,
this layer is usually the final softmax layer. For other tasks, this layer is usually
the penultimate layer for DAG and series networks and the final layer for
The default value is the penultimate layer in DAG and series networks, and the
final layer in
dlnetwork objects. Use the
analyzeNetwork function to examine your network and select the
Format — Data format
character vector | string
Data format assigning a label to each dimension of the input data, specified as a character vector or a string. Each character in the format must be one of the following dimension labels:
For more information, see
OutputUpsampling — Output upsampling method
'bicubic' (default) |
Output upsampling method, specified as the comma-separated pair consisting
'OutputUpsampling' and one of the following values:
'bicubic'— Use bicubic interpolation to produce a smooth map the same size as the input data.
'nearest'— Use nearest-neighbor interpolation to expand the map to the same size as the input data.
'none'— Use no upsampling. The map can be smaller than the input data.
'bicubic', the computed map is upsampled to the size of the input
data using the
function for 2-D data and the
imresize3 (Image Processing Toolbox) function for 3-D data. For 3-D data, the option
imresize3 with the
ExecutionEnvironment — Hardware resource
'auto' (default) |
Hardware resource for computing the map, specified as the comma-separated pair
'ExecutionEnvironment' and one of the
'auto'— Use the GPU if one is available. Otherwise, use the CPU.
'cpu'— Use the CPU.
'gpu'— Use the GPU.
The GPU option requires Parallel Computing Toolbox™.
To use a GPU for deep
learning, you must also have a supported GPU device. For information on supported devices, see
GPU Support by Release (Parallel Computing Toolbox).
If you choose the
'gpu' option and Parallel Computing Toolbox and a suitable GPU are not available, then the software returns an
scoreMap — Grad-CAM importance map
numeric matrix | numeric array
Grad-CAM importance map, returned as a numeric matrix or a numeric array. Areas in the map with higher positive values correspond to regions of input data that contribute positively to the prediction.
For classification tasks,
scoreMapis the gradient of the final classification score for the specified class, with respect to each feature in the feature layer.
For other types of tasks,
scoreMapis the gradient of the reduced output of the reduction layer, with respect to each feature in the feature layer.
scoreMap(i,j) corresponds to the Grad-CAM importance at the spatial
(i,j). If you provide
label as a vector
of categoricals, character vectors, or strings, then
corresponds to the map for
featureLayer — Name of feature layer
Name of the feature layer to extract the feature map from when computing the Grad-CAM map, returned as a string.
gradCAM chooses a feature layer to use to compute
the Grad-CAM map. This layer is the final layer with nonsingleton spatial dimensions.
You can specify which feature layer to use using the
name-value argument. When you specify the
featureLayer returns the same value.
reductionLayer — Name of reduction layer
Name of the reduction layer to extract output activations from when computing the Grad-CAM map, returned as a string.
gradCAM chooses a reduction layer to use to compute
the Grad-CAM map. This layer is the penultimate layer in DAG and series networks, and
the final layer in
dlnetwork objects. You can also specify which
reduction layer to use using the
argument. When you specify the
'ReductionLayer' name-value argument,
reductionLayer returns the same value.
Gradient-weighted class activation mapping (Grad-CAM) is an explainability technique that can be used to help understand the predictions made by a deep neural network . Grad-CAM, a generalization of the CAM technique, determines the importance of each neuron in a network prediction by considering the gradients of the target flowing through the deep network.
Grad-CAM computes the gradient of a differentiable output, for example class score, with respect to the convolutional features in the chosen layer. The gradients are spatially pooled to find the neuron importance weights. These weights are then used to linearly combine the activation maps and determine which features are most important to the prediction.
Suppose you have an image classification network with output yc, representing the score for class c, and want to compute the Grad-CAM map for a convolutional layer with k feature maps (channels), Aki,j, where i,j indexes the pixels. The neuron importance weight is
where N is the total number of pixels in the feature map. The Grad-CAM map is then a weighted combination of the feature maps with an applied ReLU:
The ReLU activation ensures you get only the features that have a positive contribution to the class of interest. The output is therefore a heatmap for the specified class, which is the same size as the feature map. The Grad-CAM map is then upsampled to the size of the input data.
Although Grad-CAM is commonly used for image classification tasks, you can compute a
Grad-CAM map for any differentiable activation. For example, for semantic segmentation
tasks, you can calculate the Grad-CAM map by replacing
yc with , where S is the set of pixels of interest and
yi,jc is 1 if pixel
(i,j) is predicted to be class c, and 0 otherwise
. You can use the
gradCAM function for nonclassification tasks by specifying a suitable
reduction function that reduces the output activations of the reduction layer to a single
value and takes the place of yc in the neuron
importance weight equation.
reductionFcnfunction receives the output from the reduction layer as a traced
dlarrayobject. The function must reduce this output to a scalar
gradCAMthen differentiates with respect to the activations of the feature layer. For example, to compute the Grad-CAM map for channel 208 of the softmax activations of a network, the reduction function is
@(x)(x(208)). This function receives the activations and extracts the 208th channel.
gradCAMfunction automatically chooses reduction and feature layers to use when computing the Grad-CAM map. For some networks, the chosen layers might not be suitable. For example, if your network has multiple layers that can be used as the feature layer, then the function chooses one of those layers, but its choice might not be the most suitable. For such networks, specify which feature layer to use using the
 Selvaraju, Ramprasaath R., Michael Cogswell, Abhishek Das, Ramakrishna Vedantam, Devi Parikh, and Dhruv Batra. “Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization.” 2017 (October 2017): 618–626, https://doi.org/10.1109/ICCV.2017.74.
 Vinogradova, Kira, Alexandr Dibrov, and Gene Myers. “Towards Interpretable Semantic Segmentation via Gradient-Weighted Class Activation Mapping.” Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 10 (April 2020): 13943–13944, https://doi.org/10.1609/aaai.v34i10.7244.
- Grad-CAM Reveals the Why Behind Deep Learning Decisions
- Explore Semantic Segmentation Network Using Grad-CAM
- Understand Network Predictions Using LIME
- Understand Network Predictions Using Occlusion
- Investigate Network Predictions Using Class Activation Mapping
- Multilabel Image Classification Using Deep Learning