how to implement CBAM with a pretarined network in matlab. i have seen tutorials doing this on other platforms but not on matlab.
20 vues (au cours des 30 derniers jours)
Afficher commentaires plus anciens
%this is the code i found on GitHub but couldn't implement it
lgraph = layerGraph();
tempLayers = imageInputLayer([227 227 3],"Name","imageinput");
lgraph = addLayers(lgraph,tempLayers);
tempLayers = maxPooling2dLayer([5 5],"Name","maxpool_1","Padding","same");
lgraph = addLayers(lgraph,tempLayers);
tempLayers = averagePooling2dLayer([5 5],"Name","avgpool2d_1","Padding","same");
lgraph = addLayers(lgraph,tempLayers);
tempLayers = [
multiplicationLayer(2,"Name","multiplication_1")
fullyConnectedLayer(10,"Name","fc_1")
fullyConnectedLayer(10,"Name","fc_2")
fullyConnectedLayer(10,"Name","fc_3")
leakyReluLayer(0.01,"Name","leakyrelu_1")];
lgraph = addLayers(lgraph,tempLayers);
tempLayers = averagePooling2dLayer([5 5],"Name","avgpool2d_2","Padding","same");
lgraph = addLayers(lgraph,tempLayers);
tempLayers = maxPooling2dLayer([5 5],"Name","maxpool_2","Padding","same");
lgraph = addLayers(lgraph,tempLayers);
tempLayers = [
multiplicationLayer(2,"Name","multiplication_2")
leakyReluLayer(0.01,"Name","leakyrelu_2")
sigmoidLayer("Name","sigmoid")
yolov2OutputLayer([16 16;32 32],"Name","yolov2-out")];
lgraph = addLayers(lgraph,tempLayers);
% 清理辅助变量
clear tempLayers;
%链接所有层
lgraph = connectLayers(lgraph,"imageinput","maxpool_1");
lgraph = connectLayers(lgraph,"imageinput","avgpool2d_1");
lgraph = connectLayers(lgraph,"maxpool_1","multiplication_1/in2");
lgraph = connectLayers(lgraph,"avgpool2d_1","multiplication_1/in1");
lgraph = connectLayers(lgraph,"leakyrelu_1","avgpool2d_2");
lgraph = connectLayers(lgraph,"leakyrelu_1","maxpool_2");
lgraph = connectLayers(lgraph,"avgpool2d_2","multiplication_2/in1");
lgraph = connectLayers(lgraph,"maxpool_2","multiplication_2/in2");
0 comments on commit b93ff1d
2 commentaires
Tian,HCong
le 27 Mai 2024
Hello, have you resolved your problem now? I am also very interested in this problem.
Réponses (0)
Voir également
Catégories
En savoir plus sur Call Python from MATLAB dans Help Center et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!