Why does tool "Detect Objects Using Deep Learning" always use CPU?

1194
2
Jump to solution
09-29-2020 09:55 AM
YunfeiXie
New Contributor

I'm using GTX 1070 as my GPU(the only graphic card I have on this computer), installed the required framework, CUDA and cuDNN. Under environments tab, I changed the Process Type to "GPU" and GPU ID is 0. However, when I hit "Run", my CPU usage goes directly to 100% whereas my GPU is doing nothing. From the documentation I'm guessing it couldn't find my GPU so it uses the CPU, but I'm very confused why this would happen.

0 Kudos
1 Solution

Accepted Solutions
DanPatterson
MVP Esteemed Contributor
2 Replies
MennoTerBals
New Contributor II

Hi @YunfeiXie ,

Did you ever find out why the tool is not using GPU?

I'm having the same issue and am looking for a way to utilize GPU.

Cheers, Menno

0 Kudos
DanPatterson
MVP Esteemed Contributor