You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I use 40GB RAM to run your code, and I also run the training part for each layer separately, but since the size of the image features is very high (especially in the initial layers), I am facing a memory error. What was the RAM of your system to run this code (featdec_deeprecon_500voxel_vgg19_allunits_fastl2lir_alpha100_train)???
If I want to reduce the number of units in each layer, how should I do it? I don't understand how you can select a single group per layer. I used a pre-trained VGG (imagenet-vgg-verydeep-19.mat) and MatConvNet for image features.
The text was updated successfully, but these errors were encountered:
I use 40GB RAM to run your code, and I also run the training part for each layer separately, but since the size of the image features is very high (especially in the initial layers), I am facing a memory error. What was the RAM of your system to run this code (featdec_deeprecon_500voxel_vgg19_allunits_fastl2lir_alpha100_train)???
If I want to reduce the number of units in each layer, how should I do it? I don't understand how you can select a single group per layer. I used a pre-trained VGG (imagenet-vgg-verydeep-19.mat) and MatConvNet for image features.
The text was updated successfully, but these errors were encountered: