nihui
nihui
> After I modified the param file, the following information appeared using ncnnoptimize. Does anyone know how to solve it? > > layer YoloV5Focus not exists or registered > 段错误...
最新版本ncnn的ncnnoptimize已经支持自定义op了
load_param 前设置 net.opt.use_vulkan_compute = true;
什么是可信选型版本？ 可以来qq群交流，群号见 ncnn github readme
You may interest in this project https://github.com/nihui/ncnn-android-nanodet
输出相差了一点点点点，这属于计算精度误差了 ``` $ python model.trace_pnnx.py tensor([[ 0.1400, -0.2828, -0.5181, 0.2307, 0.3460, 0.0774, -0.6353, -0.1958, 0.1738, -0.3608, 0.5223, 0.1877, -0.1621, 0.1208, -0.2258, -0.1601, -0.0332, -0.0391, -0.3655, -0.1871, -0.0534, -0.2243, 0.3133, 0.1827,...
针对onnx模型转换的各种问题，推荐使用最新的pnnx工具转换到ncnn In view of various problems in onnx model conversion, it is recommended to use the latest pnnx tool to convert your model to ncnn ```shell pip install pnnx pnnx...
https://github.com/Tencent/ncnn/pull/5435
implemented in https://github.com/Tencent/ncnn/commit/08b7d99a75ad70fdc5c1d9e5debc675fce5e83a9
这里有 yolov8 https://github.com/Tencent/ncnn/pull/5506