Hello,
I'm interested in testing the HRNetV2-OCR+PSA(p) with OpenCV, but I believe there is no way to load a .pth file. I'm checking how to use pytorch to load the network and then export it to ONNX, but It seems that I need to know the model definition to properly load the network. Am I going in the right direction? Sorry if the question seems naive, I don't have a lot of experience with ML, especially with pytorch.
Thanks in advance
Hello,
I'm interested in testing the HRNetV2-OCR+PSA(p) with OpenCV, but I believe there is no way to load a .pth file. I'm checking how to use pytorch to load the network and then export it to ONNX, but It seems that I need to know the model definition to properly load the network. Am I going in the right direction? Sorry if the question seems naive, I don't have a lot of experience with ML, especially with pytorch.
Thanks in advance