示例:ONNX模型的量化处理(从float32到int8) from onnxruntime.quantization import quantize_dynamic, QuantType import onnx # 加载已导出的ONNX模型 model_path = "matmul_model.onnx" quantized_model_path = "matmul_model_quantized.onnx" # 对模型进行动态量化 quantize_dynamic( model_path, quantized_mo...