Hello , I am working on using darknet yolov3 model into intel openvino with support of Intel HD graphics. I have converted the yolov3 model into Intermediete Representation using model optimizer. I loaded the model into ```IECore()``` on the GPU device. The problem is it says , the first layer named inputs is not supported by gpu. what should i do?
Link Copied
Hi Rahman,
Are you using the latest version of OpenVINO (2020.1)?
Can you please mention the command you have used to convert the model?
Also, please specify the complete error log observed while loading the model.
You may also refer to Convert YOLOv3 Model to IR for additional information.
Regards,
Jaivin
Hello. I already converted Yolov3 to IR , that is not the problem . the actual problem is , when i try to load the xml and bin files into the "GPU" device , it shows me that the GPU device doesnt support the ```inputs``` layer.. i am copying the response
model_xml="frozen_darknet_yolov3_model.xml" model_weights= model_xml[:-3]+"bin" net=IENetwork(model= model_xml,weights= model_weights) plugin= IECore() exec_net= plugin.load_network( net,device_name="GPU") supported_layers = plugin.query_network(network=net, device_name="GPU") # Check for any unsupported layers, and let the user # know if anything is missing. Exit the program, if so. unsupported_layers = [l for l in net.layers.keys() if l not in supported_layers] if len(unsupported_layers) != 0: print("Unsupported layers found: {}".format(unsupported_layers)) print("Check whether extensions are available to add to IECore.") exit(1)
output is,
Unsupported layers found: ['inputs'] Check whether extensions are available to add to IECore.
Hi,
Assuming that you are using latest version of OpenVINO (2020.1), can you test the model with a suitable demo or sample from the resources?
Does the model run on CPU? If yes, try using the Heterogeneous Plugin which could execute not supported layers on fallback devices like CPU.
Also, you can check the supported layers.
Regards,
Jaivin
```Does the model run on CPU? ```
yes. i am trying the proposed methods and informing you God willing
For more complete information about compiler optimizations, see our Optimization Notice.