search

csxmli2016 / DFDNet

Blind Face Restoration via Deep Multi-scale Component Dictionaries (ECCV 2020)
914 stars 216 forks source link

did anybody succeed in making it working on gpu without CUDA out of memory error? #49

Closed alessiapacca closed 3 years ago

alessiapacca commented 3 years ago

I was never able to run the test on a GPU, using a 2070 gpu. I was wondering if anybody was able to find where the problem is with the allocation of memory: I thought it was the number of test images, or the dictionary, but I used my own small dictionary and just 10 images, and it still gives me a cuda out of memory error.

csxmli2016 commented 3 years ago

I was never able to run the test on a GPU, using a 2070 gpu. I was wondering if anybody was able to find where the problem is with the allocation of memory: I thought it was the number of test images, or the dictionary, but I used my own small dictionary and just 10 images, and it still gives me a cuda out of memory error.

Hi, only 8GB memory in 2070 is not enough to run this code on GPU mode. You can use 1080Ti or 2080 Ti (at least 12 GB).