I can’t deal with my Post-traing-static-Quantization model.
It came across the detection problem, I can’t get the right detection outcome after applying Pytorch FX-mode static quantization.
It seems that the accuracy is 0 after I using the Quantized YOLOv7 to do inference job. (Didn’t find the object location? )
What may the error come from ? Or how do I to test where the probably error come from?
I am using Pytorch 2.1.0 ，
Using YOLOv7 & YOLOv7-tiny model to doing my cumtom dataset training and inference job.