Closed MrMzl closed 1 year ago
开发者朋友你好,您给出的c++ demo的日志链接(提取码: yy4w)不对,请您检查一下。其次您这个模型是哪里提供的,rk官方提供的吗
郑工您好,下面是新上传的c++ demo日志链接:https://pan.baidu.com/s/1iCQDm2bCjcX25zuKSFKgvA 提取码:qhu0 另外,链接https://pan.baidu.com/s/1JDvgZnPWruSRNdhMO1oYkg 提取码: yy4w 中的模型为转换后的模型(参考链接https://github.com/airockchip/rknn_model_zoo/tree/main/models/CV/object_detection/yolo/RKNN_model_convert)
郑工您好,下面是新上传的c++ demo日志链接:https://pan.baidu.com/s/1iCQDm2bCjcX25zuKSFKgvA 提取码:qhu0 另外,链接https://pan.baidu.com/s/1JDvgZnPWruSRNdhMO1oYkg 提取码: yy4w 中的模型为转换后的模型(参考链接https://github.com/airockchip/rknn_model_zoo/tree/main/models/CV/object_detection/yolo/RKNN_model_convert)
参考链接失效了
你这个模型是自己的模型吗?RK官方放在百度网盘的模型你有测试过吗?
你这个模型是自己的模型吗?RK官方放在百度网盘的模型你有测试过吗?
这个模型是利用rk官方提供的转换脚本convert_yolo_ppyolo.sh和模型文件yolov5s_relu_tk2.pt进行转换得到的,配置文件及运行日志见链接: https://pan.baidu.com/s/1EBeEKC6ffppDdnV0YQ0Xxw 提取码: 4i2c
我刚才用官方提供的转换后的模型进行了重新测试,同样会出现相同的错误“Segmentation fault”,模型链接https://eyun.baidu.com/enterprise/share/link?cid=8272257679089781337&uk=2751701137&sid=202211118572878233#sharelink/path=%2Frknn_model_zoo%2Fmodels%2FCV%2Fobject_detection%2Fyolo%2Fyolov5%2Fdeploy_models%2Ftoolkit2%2Fmodel_cvt%2FRK356X&parent_path=%2FGPU
你这个模型是自己的模型吗?RK官方放在百度网盘的模型你有测试过吗?
这个模型是利用rk官方提供的转换脚本convert_yolo_ppyolo.sh和模型文件yolov5s_relu_tk2.pt进行转换得到的,配置文件及运行日志见链接: https://pan.baidu.com/s/1EBeEKC6ffppDdnV0YQ0Xxw 提取码: 4i2c 我刚才用官方提供的转换后的模型进行了重新测试,同样会出现相同的错误“Segmentation fault”,模型链接https://eyun.baidu.com/enterprise/share/link?cid=8272257679089781337&uk=2751701137&sid=202211118572878233#sharelink/path=%2Frknn_model_zoo%2Fmodels%2FCV%2Fobject_detection%2Fyolo%2Fyolov5%2Fdeploy_models%2Ftoolkit2%2Fmodel_cvt%2FRK356X&parent_path=%2FGPU
输入以下命令,会出现什么结果呢?
ldd 你编译出来的程序
你这个模型是自己的模型吗?RK官方放在百度网盘的模型你有测试过吗?
这个模型是利用rk官方提供的转换脚本convert_yolo_ppyolo.sh和模型文件yolov5s_relu_tk2.pt进行转换得到的,配置文件及运行日志见链接: https://pan.baidu.com/s/1EBeEKC6ffppDdnV0YQ0Xxw 提取码: 4i2c 我刚才用官方提供的转换后的模型进行了重新测试,同样会出现相同的错误“Segmentation fault”,模型链接https://eyun.baidu.com/enterprise/share/link?cid=8272257679089781337&uk=2751701137&sid=202211118572878233#sharelink/path=%2Frknn_model_zoo%2Fmodels%2FCV%2Fobject_detection%2Fyolo%2Fyolov5%2Fdeploy_models%2Ftoolkit2%2Fmodel_cvt%2FRK356X&parent_path=%2FGPU
输入以下命令,会出现什么结果呢?
ldd 你编译出来的程序
firefly@firefly:~/FastDeploy-release-1.0.3/examples/vision/detection/rkyolo/cpp/build/install$ ldd infer_rkyolo linux-vdso.so.1 (0x0000007fbca8f000) libfastdeploy.so.1.0.3 => ./lib/libfastdeploy.so.1.0.3 (0x0000007fbbe82000) libonnxruntime.so.1.12.0 => ./lib/libonnxruntime.so.1.12.0 (0x0000007fbb0ec000) librknnrt.so => ./lib/librknnrt.so (0x0000007fbad74000) libopencv_dnn.so.3.4 => ./lib/libopencv_dnn.so.3.4 (0x0000007fba987000) libopencv_highgui.so.3.4 => ./lib/libopencv_highgui.so.3.4 (0x0000007fba96d000) libopencv_ml.so.3.4 => ./lib/libopencv_ml.so.3.4 (0x0000007fba8dd000) libopencv_objdetect.so.3.4 => ./lib/libopencv_objdetect.so.3.4 (0x0000007fba85b000) libopencv_shape.so.3.4 => ./lib/libopencv_shape.so.3.4 (0x0000007fba82c000) libopencv_stitching.so.3.4 => ./lib/libopencv_stitching.so.3.4 (0x0000007fba7be000) libopencv_superres.so.3.4 => ./lib/libopencv_superres.so.3.4 (0x0000007fba794000) libopencv_videostab.so.3.4 => ./lib/libopencv_videostab.so.3.4 (0x0000007fba751000) libpaddle2onnx.so.1.0.5 => ./lib/libpaddle2onnx.so.1.0.5 (0x0000007fba195000) libopencv_calib3d.so.3.4 => ./lib/libopencv_calib3d.so.3.4 (0x0000007fba08a000) libopencv_features2d.so.3.4 => ./lib/libopencv_features2d.so.3.4 (0x0000007fb9fe5000) libopencv_flann.so.3.4 => ./lib/libopencv_flann.so.3.4 (0x0000007fb9f81000) libopencv_photo.so.3.4 => ./lib/libopencv_photo.so.3.4 (0x0000007fb9eef000) libopencv_video.so.3.4 => ./lib/libopencv_video.so.3.4 (0x0000007fb9e9c000) libopencv_videoio.so.3.4 => ./lib/libopencv_videoio.so.3.4 (0x0000007fb9e5d000) libopencv_imgcodecs.so.3.4 => ./lib/libopencv_imgcodecs.so.3.4 (0x0000007fb9c95000) libopencv_imgproc.so.3.4 => ./lib/libopencv_imgproc.so.3.4 (0x0000007fb989a000) libopencv_core.so.3.4 => ./lib/libopencv_core.so.3.4 (0x0000007fb958d000) libstdc++.so.6 => /usr/lib/aarch64-linux-gnu/libstdc++.so.6 (0x0000007fb93e1000) libm.so.6 => /lib/aarch64-linux-gnu/libm.so.6 (0x0000007fb9324000) libgcc_s.so.1 => /lib/aarch64-linux-gnu/libgcc_s.so.1 (0x0000007fb9300000) libc.so.6 => /lib/aarch64-linux-gnu/libc.so.6 (0x0000007fb918e000) libdl.so.2 => /lib/aarch64-linux-gnu/libdl.so.2 (0x0000007fb917a000) librt.so.1 => /lib/aarch64-linux-gnu/librt.so.1 (0x0000007fb9162000) libpthread.so.0 => /lib/aarch64-linux-gnu/libpthread.so.0 (0x0000007fb9133000) /lib/ld-linux-aarch64.so.1 (0x0000007fbca61000) libz.so.1 => /lib/aarch64-linux-gnu/libz.so.1 (0x0000007fb9106000)
我花点时间排查下是什么问题,我这边没有复现。不好意思,您稍等一下哈
没有发现你这个情况,群里的网友也是用的这个模型,编译后能正常使用
郑工您好,下面是新上传的c++ demo日志链接:https://pan.baidu.com/s/1iCQDm2bCjcX25zuKSFKgvA 提取码:qhu0 另外,链接https://pan.baidu.com/s/1JDvgZnPWruSRNdhMO1oYkg 提取码: yy4w 中的模型为转换后的模型(参考链接https://github.com/airockchip/rknn_model_zoo/tree/main/models/CV/object_detection/yolo/RKNN_model_convert)
找到Bug了,你这里执行的命令行,应该是 ./infer model_path image_path,而不是模型的文件夹
郑工您好,下面是新上传的c++ demo日志链接:https://pan.baidu.com/s/1iCQDm2bCjcX25zuKSFKgvA 提取码:qhu0 另外,链接https://pan.baidu.com/s/1JDvgZnPWruSRNdhMO1oYkg 提取码: yy4w 中的模型为转换后的模型(参考链接https://github.com/airockchip/rknn_model_zoo/tree/main/models/CV/object_detection/yolo/RKNN_model_convert)
找到Bug了,你这里执行的命令行,应该是 ./infer model_path image_path,而不是模型的文件夹
您好,郑工,按您这边的意见修改后能够正常推理(Python、C++均可),感谢您的指导。
环境
运行FastDeploy-release-1.0.3/examples/vision/detection/rkyolo/cpp中的demo
运行FastDeploy-release-1.0.3/examples/vision/detection/rkyolo/python中的demo
环境编译日志
https://pan.baidu.com/s/1R9kIj34-0qPdTK86uCnRMw 提取码: eeba