Why is it said not enough input arguments at momentum1 = zeros(size(W1)); ?
function [W1, W5, Wo] = MnistConv(W1, W5, Wo, X, D)
alpha = 0.01;
beta = 0.95;
momentum1 = zeros(size(W1));
momentum5 = zeros(size(W5));
momentumo = zeros(size(Wo));
N = length(D);
bsize = 100;
blist = 1:bsize:(N-bsize+1);
for batch = 1:length(blist)
dW1 = zeros(size(W1));
dW5 = zeros(size(W5));
dWo = zeros(size(Wo));
begin = blist(batch);
for k = begin:begin+bsize-1
x = X(:, :, k);
y1 = Conv(x, W1);
y2 = ReLU(y1);
y3 = Pool(y2);
y4 = reshape(y3, [],1);
v5 = W5*y4;
y5 = ReLU(v5);
v = Wo*y5;
y = Softmax(v);
d = zeros(10, 1);
d(sub2ind(size(d), D(k), 1)) = 1;
e = d - y;
delta = e;
e5 = Wo' * delta;
delta5= (y5 > 0).* e5;
e4 = W5' * delta5;
e3 = reshape(e4, size(y3));
e2 = zeros(size(y2));
W3 = ones(size(y2)) / (2*2);
for c = 1:20
e2(:, :,c) = kron(e3(:, :, c), ones([2 2])) .*W3(:, :, c);
end
delta2 = (y2 > 0) .* e2;
delta1_x = zeros(size(W1));
for c = 1:20
delta1_x(:, :,c) = conv2(x(:, :), rot90(delta2(:, :,c), 2), 'valid');
end
dW1 = dW1 + delta1_x;
dW5 = dW5 + delta5*y4';
dWo = dWo + delta*y5';
end
dW1 = dW1 / bsize;
dW5 = dW5 / bsize;
dWo = dWo / bsize;
momentum1 = alpha*dW1 + beta*momentum1;
W1 = W1 + momentum1;
momentum5 = alpha*dW5 + beta*momentum5;
W5 = W5 + momentum5;
momentumo = alpha*dWo + beta*momentumo;
Wo = Wo + momentumo;
end
end

 채택된 답변

Stephen23
Stephen23 2020년 10월 25일

0 개 추천

"Why is it said not enough input arguments at momentum1 = zeros(size(W1)); ?"
Because you called the function without any input arguments (probably by clicking the big greeen "Run" button).
You need to call your function (e.g. from the MATLAB command line) with five input arguments, e,g.:
W1 = 1;
W5 = 2;
Wo = 3;
X = 4;
D = 5;
[W1, W5, Wo] = MnistConv(W1, W5, Wo, X, D)
Of course you will need to define appropriate arrays and values to suit whatever task you are doing, not just copy 1, 2, 3, etc. from my answer. And in future ignore that big green button.

댓글 수: 1

jikwan Bae
jikwan Bae 2020년 12월 2일
it's not test code. run test MnistConvcode then run Matlab.

댓글을 달려면 로그인하십시오.

추가 답변 (0개)

카테고리

도움말 센터File Exchange에서 Performance and Memory에 대해 자세히 알아보기

태그

질문:

2020년 10월 25일

댓글:

2020년 12월 2일

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by