function cnn = cnn_initialize(cnn)
%CNN_INIT initialize the weights and biases, and other parameters
%
index = 0;
num_layer = numel(cnn.layer);
for in = 1:num_layer
switch cnn.layer{in}{1}
case 'input'
index = index + 1;
height = cnn.layer{in}{3}(1);
width = cnn.layer{in}{3}(2);
mini_size = cnn.layer{in}{2};
cnn.weights{index} = [];
cnn.biases{index} = [];
cnn.nabla_w{index} = [];
cnn.nabla_b{index} = [];
%n*n*m
cnn.a{index} = [];
cnn.z{index} = [];
cnn.delta{index} = [];
cnn.mini_size = mini_size;
case 'conv'
index = index + 1;
%kernel height, width, number
ker_height = cnn.layer{in}{3}(1);
ker_width = cnn.layer{in}{3}(2);
ker_num = cnn.layer{in}{2};
cnn.weights{index} = grand(ker_height,ker_width,ker_num) - 0.5;
cnn.biases{index} = grand(1,ker_num) - 0.5;
cnn.nabla_w{index} = zeros(ker_height,ker_width,ker_num);
cnn.nabla_b{index} = zeros(1,ker_num);
height = height - ker_height + 1;
width = width - ker_width + 1;
cnn.a{index} = zeros(height,width,mini_size,ker_num);
cnn.z{index} = zeros(height,width,mini_size,ker_num);
cnn.delta{index} = zeros(height,width,mini_size,ker_num);
case 'pool'
index = index + 1;
%kernel height, width, number
ker_height = cnn.layer{in}{3}(1);
ker_width = cnn.layer{in}{3}(2);
cnn.weights{index} = [];
cnn.biases{index} = [];
cnn.nabla_w{index} = [];
cnn.nabla_b{index} = [];
height = height / ker_height;
width = width / ker_width;
cnn.a{index} = zeros(height,width,mini_size,ker_num);
cnn.z{index} = [];
cnn.delta{index} = zeros(height,width,mini_size,ker_num);
case 'flat'
index = index + 1;
cnn.weights{index} = [];
cnn.biases{index} = [];
cnn.nabla_w{index} = [];
cnn.nabla_b{index} = [];
cnn.a{index} = zeros(height*width*ker_num,mini_size);
cnn.z{index} = [];
cnn.delta{index} = zeros(height*width*ker_num,mini_size);
case 'full'
index = index + 1;
%kernel height, width, number
neuron_num = cnn.layer{in}{2};
neuron_num0 = size(cnn.a{in-1},1);
cnn.weights{index} = grand(neuron_num,neuron_num0) - 0.5;
cnn.biases{index} = grand(neuron_num,1) - 0.5;
cnn.nabla_w{index} = zeros(neuron_num,neuron_num0);
cnn.nabla_b{index} = zeros(neuron_num,1);
cnn.a{index} = zeros(neuron_num,mini_size);
cnn.z{index} = zeros(neuron_num,mini_size);
cnn.delta{index} = zeros(neuron_num,mini_size);
case 'output'
index = index + 1;
%kernel height, width, number
neuron_num = cnn.layer{in}{2};
neuron_num0 = size(cnn.a{in-1},1);
cnn.weights{index} = grand(neuron_num,neuron_num0) - 0.5;
cnn.biases{index} = grand(neuron_num,1);
cnn.nabla_w{index} = zeros(neuron_num,neuron_num0);
cnn.nabla_b{index} = zeros(neuron_num,1);
cnn.a{index} = zeros(neuron_num,mini_size);
cnn.z{index} = zeros(neuron_num,mini_size);
cnn.delta{index} = zeros(neuron_num,mini_size);
otherwise
end
end
end
CNN和BP MNIST手写体数字识别.7z
需积分: 0 53 浏览量
2021-03-31
11:02:21
上传
评论 22
收藏 21.24MB 7Z 举报
CNN和BP MNIST手写体数字识别.7z （44个子文件） 
MNISTMatlab程序CNN 和 BP MNISTMatlab程序 data 
mnist.mat 22.12MB fig cnn 
cnnresult.png 22KB bp relu-z2-m100-p5000-30-30.png 13KB a100-100-m-100-me-50000-e-1-l-5.png 62KB a80-80-m-100-me-50000-e-1-l-5.png 27KB a120-120-m-100-me-50000-e-1-l-5.png 27KB relu-z2-m100-p5000.png 13KB cnn softmax.m 189B cnn_feedforward.m 1KB readme.m 75B showdigit.m 400B gzeros.m 165B relu.m 81B run.m 1KB cnn_kron.m 297B sigmoid.m 72B cnn_backpropagation2.m 3KB grandn.m 168B showWeights.m 1KB gones.m 159B relu_prime.m 119B cnn_evaluate.m 445B cnn_backpropagation.m 3KB sigmoid_prime.m 95B copyToGPU.m 142B loadMnistDataScript.m 252B cnn_initialize.m 4KB grand.m 165B bp softmax.m 188B acti_sigmoid.m 77B readme.m 601B showdigit.m 400B networksmnist.m 1KB evaluatemnist.m 425B bp_backpropagation.m 812B networksmnist2.m 2KB feedforward.m 476B acti_relu_prime.m 124B loadMnistDataScript.m 255B bp_feedforward.m 475B networksmnist3.m 2KB SGD.m 644B acti_relu.m 86B acti_sigmoid_prime.m 105B
评论0