search

sdemyanov / ConvNet

Convolutional Neural Networks for Matlab for classification and segmentation, including Invariang Backpropagation (IBP) and Adversarial Training (AT) algorithms. Trained on GPU, require cuDNN v5.
240 stars 141 forks source link

Performance changes #6

Closed rmanor closed 10 years ago

rmanor commented 10 years ago

btw, using matrices instead of cells is much faster and uses less memory..

sdemyanov commented 10 years ago

You are probably right, but I don't have time to change it right now. If you do it, I'll accept your request. That should not be too difficult.