0

Is there any way I can pass existing images in my system through a trained VGG with torch? I am using Ubuntu 14.04 and unfortunately do not have a GPU. I have searched quite extensively but all the ones I have found require a GPU. Are there other ways to use VGG without torch? I'm open to suggestions but the method should not require a GPU.

stop-cran
  • 4,229
  • 2
  • 30
  • 47
Sibi
  • 2,221
  • 6
  • 24
  • 37

1 Answers1

0

While running the network using a GPU will make things a lot faster, you can run the network in CPU mode only.

Once you load the model, pretrained on a GPU you can simple convert it to CPU as follow:

model = model:float()

You can easily load an image from your computer with the help of the image library and then do a forward pass

local img = image.load(imagefile,3,'byte') local output = model:forward(img)

A B
  • 497
  • 2
  • 9
  • Thanks a lot! Which VGG model are you talking about and where can I find it? – Sibi Jul 06 '16 at 11:22
  • @Sibi You can find here a list of the available pretrained networks: https://github.com/torch/torch7/wiki/ModelZoo (they are caffe models), and you can load them as a torch model using loadcaffe module as described in the link from above. – A B Jul 06 '16 at 23:52