Closed zhangliliang closed 10 years ago
Sure the SVHN/Cifar-100 configurations will be released after CVPR deadline.
For MNIST the results are achieved using Theano, in a fairly standard architecture. You can refer the paper for the specific parameter settings.
Hi, I reproduce the experiment and appreciate that DSN do a great job. I try a idea to improve the SVM and get a bit improvement on CIFAR10. I am curious about whether it would work on other dataset under DSN framework. Does the author has a future plan to release other dataset's(MNIST, SVHN, etc.) configurations?