bazilas / matconvnet-resnet Goto Github PK

Hello, Thank you very much for the code.
I am running it on the Matlab 2015b.
I am getting this error

Do you have any idea what's the issue?

Hi all.I have a question about the
opts.learningRate = [0.01ones(1,3) 0.1ones(1,80) 0.01ones(1,10) 0.001ones(1,20)] ;
As I known,if the learning rate is small,it is easy to overfit.So I think the
opts.learningRate = [0.01ones(1,3) 0.1ones(1,80) 0.01ones(1,20) 0.001ones(1,10)] ; is better.

val: epoch 01: 79/ 79: 26.5 (27.3) Hz error: 0.68100 objective: 1.80163
Error using checkArgsForHandleToPrint
Handle input argument contains nonhandle values.

Error in checkArgsForHandleToPrint

Error in print>LocalCreatePrintJob (line 216)
handles = checkArgsForHandleToPrint(0, varargin{:});

Error in print (line 38)
[pj, inputargs] = LocalCreatePrintJob(varargin{:});

Error in cnn_train_dag (line 142)
print(1, modelFigPath, '-dpdf') ;

Error in cnn_cifar (line 79)
[net, info] = cnn_train_dag(net, imdb, getBatch(opts), ...

any suggestions...

Hi Bazilas, thanks for sharing us the great work. May I know if any script used to predict a new image using the trained reset model is provided? Thanks

Hi,

Thank you for the implementation. This helps a lot.
Is there pretrained model with a mat file, for both cifar 10 and 100?
It will be good to save the computation cost.

Thank you.