Gary-zyt
commented
3 months ago 能否在runtime_option中设定调用多个npu？官方文档中我只看到，选择调用npu或cpu，但目前调用npu发现只能调用单个的（Core0） model = fd.vision.segmentation.PaddleSegModel( model_file, params_file, config_file, runtime_option=runtime_option, model_format=fd.ModelFormat.RKNN)
Jiang-Jia-Jun
在瑞芯微rk3588上部署paddleseg的模型（如pp-liteseg），发现在npu只调用Core0，且占用率只有5%-15%左右，有没有办法提高npu的占用率或者是多核调用？python语言下,预测代码如下所示： import fastdeploy as fd import cv2 import os
def parse_arguments(): import argparse import ast parser = argparse.ArgumentParser() parser.add_argument( "--model_file", required=True, help="Path of PaddleSeg model.") parser.add_argument( "--config_file", required=True, help="Path of PaddleSeg config.") parser.add_argument( "--image", type=str, required=True, help="Path of test image file.") return parser.parse_args()
def build_option(args): option = fd.RuntimeOption() option.use_rknpu2() return option
args = parse_arguments()
setup runtime
runtime_option = build_option(args) model_file = args.model_file params_file = "" config_file = args.config_file model = fd.vision.segmentation.PaddleSegModel( model_file, params_file, config_file, runtime_option=runtime_option, model_format=fd.ModelFormat.RKNN)
model.preprocessor.disable_normalize() model.preprocessor.disable_permute()
predict
im = cv2.imread(args.image) result = model.predict(im) print(result)
visualize
vis_im = fd.vision.vis_segmentation(im, result, weight=0.5) cv2.imwrite("vis_img.png", vis_im)