Merge pull request #3932 from chrismuzyn/master

When using the openvino backend, do not look for an nvidia gpu.

modules/devices.py

@@ -183,6 +183,8 @@ def get_cuda_device_string():
 
 
 def get_optimal_device_name():
+    if backend == 'openvino':
+        return "cpu"
     if cuda_ok or backend == 'directml':
         return get_cuda_device_string()
     if has_mps() and backend != 'openvino':