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

GPU Convolution Calculation #27

Open YFje opened 8 years ago

YFje commented 8 years ago

HI

I want to use convolution layers to do feature extraction. That means after training, I can get the result from each layers.Could you tell me how to do this work base on gpu function in this code.

sdemyanov commented 8 years ago

Hi,

If you just need to look at the filters after training, you can use the Matlab functions getweights/setweights to transform the vector of parameters to layers and backward. If you want to see which images activate particular neurons, you need to backpropagate activations using operations of deconvolution and upsampling. This is not implemented in my toolbox yet.

Regards, Sergey.

On Thu, Dec 3, 2015 at 3:14 AM, YFje notifications@github.com wrote:

HI

I want to use convolution layers to do feature extraction. That means after training, I can get the result from each layers.Could you tell me how to do this work base on gpu function in this code.

— Reply to this email directly or view it on GitHub https://github.com/sdemyanov/ConvNet/issues/27.