Pytorch example of Faster R-CNN for CPU

1

Hi all,
I’m quite new to Pytorch and i’d like to find an example of Faster R-CNN that can run on a simple CPU. I’ve found several examples in GitHub but i don’t know if running them on a CPU will be very complicated. Has anyone tried? Any kind of guide would be appreciated.
Thanks in advance for your help!!

2

Just get rid of all .cuda() calls and run the script. This prevents moving the model or tensors onto a GPU.
Often there are optional arguments for the main script where you can specify using a GPU or not.

EDIT: This code uses an argument for GPU usage. (Haven’t tried it myself)

1 Like
3

Thanks @ptrblck !! This is the kind of help that i needed. i will try and came back to edit the post