examples/yolov8.cpp 按步骤操作 yolov8.2.99 导出onnx模型，测试检测结果异常

[zqqq]

error log | 日志或报错信息 | ログ

[0 Intel(R) UHD Graphics] queueC=0[1] queueG=0[1] queueT=0[1] [0 Intel(R) UHD Graphics] bugsbn1=0 bugbilz=0 bugcopc=0 bugihfa=0 [0 Intel(R) UHD Graphics] fp16-p/s/u/a=1/1/1/1 int8-p/s/u/a=1/1/1/1 [0 Intel(R) UHD Graphics] subgroup=32 basic/vote/ballot/shuffle=1/1/1/1 [0 Intel(R) UHD Graphics] fp16-8x8x16/16x8x8/16x8x16/16x16x16=0/0/0/0 [1 Intel(R) UHD Graphics] queueC=0[1] queueG=0[1] queueT=0[1] [1 Intel(R) UHD Graphics] bugsbn1=0 bugbilz=0 bugcopc=0 bugihfa=0 [1 Intel(R) UHD Graphics] fp16-p/s/u/a=1/1/1/1 int8-p/s/u/a=1/1/1/1 [1 Intel(R) UHD Graphics] subgroup=32 basic/vote/ballot/shuffle=1/1/1/1 [1 Intel(R) UHD Graphics] fp16-8x8x16/16x8x8/16x8x16/16x16x16=0/0/0/0 0 = 0.86426 at 1.87 3.34 -1.87 x -3.34 0 = 0.85693 at 7.67 11.36 -7.67 x -11.36 0 = 0.85107 at 7.29 9.34 -7.29 x -9.34 0 = 0.83984 at 7.37 8.02 -7.37 x -8.02 0 = 0.82520 at 1.37 2.14 -1.37 x -2.14 0 = 0.82031 at 0.00 0.00 2.82 x 4.96 0 = 0.79883 at 1.78 3.53 -1.78 x -3.53 0 = 0.79150 at 1.68 2.21 -1.68 x -2.21 0 = 0.69824 at 7.92 12.90 -7.92 x -12.90 32 = 0.69238 at 0.81 0.51 -0.81 x -0.51 32 = 0.55273 at 4.89 3.62 -4.89 x -3.62 32 = 0.54980 at 0.93 1.20 -0.93 x -1.20 32 = 0.54004 at 0.83 0.79 -0.83 x -0.79 32 = 0.49561 at 5.09 5.03 -5.09 x -5.03 32 = 0.47583 at 0.00 0.00 2.31 x 1.99 32 = 0.45386 at 4.80 2.35 -4.80 x -2.35

context | 编译/运行环境 | バックグラウンド

win10 x64

how to reproduce | 复现步骤 | 再現方法

1. model = YOLO("yolov8n.pt") # load an official pretrained weight model model.export(format="ncnn", dynamic=True, save_dir="/", simplify=True)
2. 编译并运行 yolov8.exe sb.png

是否是因为版本问题？