Showing results for 
Search instead for 
Did you mean: 

Does support quantized onnx models?

Associate II

I used a quantization tool called ppq to quantize my onnx model because the activation values in the middle layer of this model were too large and I needed to quantize it from float32 type to int8 type. By the way, this model was defined and trained by me using pytorch. I used STM32CUBEMX and CUBEAI version 8.1.0 to try to analyze the model and the progress bar would get stuck referring to the fact that it loaded from 0 percent to 100 percent and then went back to 0 percent, and if I didn't interrupt it he would keep going. Can anyone help me?