layumi / 2016_gan_matlab Goto Github PK

%GD
if( opts.flag2 && net.vars(net.getVarIndex('top1err')).value/batchSize<0.2)
fprintf('updating G\n');
opts.flag1 = true;
opts.flag2 = false;
end
if( opts.flag1 && net.vars(net.getVarIndex('top1err')).value/batchSize>0.6)
fprintf('updating D\n');
end
opts.flag1 = false;
opts.flag2 = true;

In any case, opts.flag1 will be false, and opts.flag2 will be true. Hence, what will be the effort of the opts.flag1 and opts.flag2 ?
Thanks

Thanks for your response. I can run the train_gan_3.m, however, when I run the test_gan_3, there is a error:

test_gan_3
错误使用 vl_nnconv
FILTERS are larger than the DATA (including padding).
出错 dagnn.Conv/forward (line 11)
outputs{1} = vl_nnconv(...
出错 dagnn.Layer/forwardAdvanced (line 85)
outputs = obj.forward(inputs, {net.params(par).value}) ;
出错 dagnn.DagNN/eval (line 88)
obj.layers(l).block.forwardAdvanced(obj.layers(l)) ;
出错 test_gan_3 (line 11)
net1.eval({'data_rand',(im)});

I don't modify any parameters. Can you run the test_gan_3 ?

Hello, when I run the train_gan_3, there is a error:

train_gan_3
11 net = GDnet_3();
Undefined variable "dagnn" or class "dagnn.GausianNoise".

Hi, Could this matlab code be runned in the model of cpu ?
Thank you.