- 首先在 VGG Convolutional Neural Networks Practical 官网上做了四个练习。现在代码可以直接用
- 但是在using pretrained models中有个错,net模型不一致,待解决!直接复制代码有问题;但是下载数据后运行Exe05是可以运行的。
net = load('data/imagenet-vgg-verydeep-16.mat') ; vl_simplenn_display(net) ; % obtain and preprocess an image im = imread('peppers.png') ; im_ = single(im) ; % note: 255 range % im_ = imresize(im_, net.normalization.imageSize(1:2)) ; % im_ = im_ - net.normalization.averageImage ; im_ = imresize(im_, net.meta.normalization.imageSize(1:2)) ; im_ = im_ - net.meta.normalization.averageImage ; % 操作的数组维度必须匹配 % run the CNN res = vl_simplenn(net, im_) ; % show the classification result scores = squeeze(gather(res(end).x)) ; [bestScore, best] = max(scores) ; figure(1) ; clf ; imagesc(im) ; title(sprintf('%s (%d), score %.3f',... net.classes.description{best}, best, bestScore)) ;
- 一、卷积
Y = VL_NNCONV(X, F, B)计算图像堆x的卷积,F是卷积核,B是偏置。X=H*W*D*N, (H,W)是图像的高和宽,D是图像深度(特征频道的数目,例彩色就是3),