Commit Graph

5 Commits

Author SHA1 Message Date
Shivam Kumar
9c1ea4fa01
Fix GPU not being used even when it available
On lots of systems onnxruntime doesn't detect GPU unless pytorch is imported before it. So despite having CUDA and CUDNN setup correctly, it is only using CPUExecutionProvider. By importing pytorch first, the issue is fixed. So let's use this until any official solution is available.

See: https://stackoverflow.com/questions/75294639/onnxruntime-inference-with-cudnn-on-gpu-only-working-if-pytorch-imported-first
2023-05-31 09:20:41 +05:30
henryruhs
dc0653ad39 Restore globals, add process time for better comparison 2023-05-30 01:03:43 +02:00
henryruhs
54f800dd0e Add GPU support, Quit on missing model, Remove globals (sorry) 2023-05-30 00:40:02 +02:00
Somdev Sangwan
4bb6ea416e reverse providers logic 2023-05-29 23:10:52 +05:30
Somdev Sangwan
32ebd9191c bruh moment 3 2023-05-29 19:13:55 +05:30