it compiled suc,but it can not run,the error is
mobileNetSSDTensorRT: NvPluginSSD.cu:795: virtual void nvinfer1::plugin::DetectionOutput::configure(const nvinfer1::Dims*, int, const nvinfer1::Dims*, int, int): Assertion `numPriorsnumLocClasses4 == inputDims[param.inputOrder[0]].d[0]' failed.
已放弃 (核心已转储)
can you tell me why?thanks very much

Using Nvidia Jetson TX2, fresh install, with these packages

ii  cuda-command-line-tools-9-0                9.0.252-1                                     arm64        CUDA command-line tools
ii  cuda-core-9-0                              9.0.252-1                                     arm64        CUDA core tools
ii  cuda-cublas-9-0                            9.0.252-1                                     arm64        CUBLAS native runtime libraries
ii  cuda-cublas-dev-9-0                        9.0.252-1                                     arm64        CUBLAS native dev links, headers
ii  cuda-cudart-9-0                            9.0.252-1                                     arm64        CUDA Runtime native Libraries
ii  cuda-cudart-dev-9-0                        9.0.252-1                                     arm64        CUDA Runtime native dev links, headers
ii  cuda-cufft-9-0                             9.0.252-1                                     arm64        CUFFT native runtime libraries
ii  cuda-cufft-dev-9-0                         9.0.252-1                                     arm64        CUFFT native dev links, headers
ii  cuda-curand-9-0                            9.0.252-1                                     arm64        CURAND native runtime libraries
ii  cuda-curand-dev-9-0                        9.0.252-1                                     arm64        CURAND native dev links, headers
ii  cuda-cusolver-9-0                          9.0.252-1                                     arm64        CUDA solver native runtime libraries
ii  cuda-cusolver-dev-9-0                      9.0.252-1                                     arm64        CUDA solver native dev links, headers
ii  cuda-cusparse-9-0                          9.0.252-1                                     arm64        CUSPARSE native runtime libraries
ii  cuda-cusparse-dev-9-0                      9.0.252-1                                     arm64        CUSPARSE native dev links, headers
ii  cuda-documentation-9-0                     9.0.252-1                                     arm64        CUDA documentation
ii  cuda-driver-dev-9-0                        9.0.252-1                                     arm64        CUDA Driver native dev stub library
ii  cuda-libraries-dev-9-0                     9.0.252-1                                     arm64        CUDA Libraries 9.0 development meta-package
ii  cuda-license-9-0                           9.0.252-1                                     arm64        CUDA licenses
ii  cuda-misc-headers-9-0                      9.0.252-1                                     arm64        CUDA miscellaneous headers
ii  cuda-npp-9-0                               9.0.252-1                                     arm64        NPP native runtime libraries
ii  cuda-npp-dev-9-0                           9.0.252-1                                     arm64        NPP native dev links, headers
ii  cuda-nvgraph-9-0                           9.0.252-1                                     arm64        NVGRAPH native runtime libraries
ii  cuda-nvgraph-dev-9-0                       9.0.252-1                                     arm64        NVGRAPH native dev links, headers
ii  cuda-nvml-dev-9-0                          9.0.252-1                                     arm64        NVML native dev links, headers
ii  cuda-nvrtc-9-0                             9.0.252-1                                     arm64        NVRTC native runtime libraries
ii  cuda-nvrtc-dev-9-0                         9.0.252-1                                     arm64        NVRTC native dev links, headers
ii  cuda-repo-l4t-9-0-local                    9.0.252-1                                     arm64        cuda repository configuration files
ii  cuda-samples-9-0                           9.0.252-1                                     arm64        CUDA example applications
ii  cuda-toolkit-9-0                           9.0.252-1                                     arm64        CUDA Toolkit 9.0 meta-package
ii  libcudnn7                                  7.1.5.14-1+cuda9.0                            arm64        cuDNN runtime libraries
ii  libcudnn7-dev                              7.1.5.14-1+cuda9.0                            arm64        cuDNN development libraries and headers
ii  libcudnn7-doc                              7.1.5.14-1+cuda9.0                            arm64        cuDNN documents and samples
ii  libnvinfer-dev                             4.1.3-1+cuda9.0                               arm64        TensorRT development libraries and headers
ii  libnvinfer-samples                         4.1.3-1+cuda9.0                               arm64        TensorRT samples and documentation
ii  libnvinfer4                                4.1.3-1+cuda9.0                               arm64        TensorRT runtime libraries
ii  tensorrt                                   4.0.2.0-1+cuda9.0                             arm64        Meta package of TensorRT

and with these defines

#define NV_TENSORRT_MAJOR 4 //!< TensorRT major version.
#define NV_TENSORRT_MINOR 0 //!< TensorRT minor version.
#define NV_TENSORRT_PATCH 2 //!< TensorRT patch version.
#define NV_TENSORRT_BUILD 0 //!< TensorRT build number.

#define NV_TENSORRT_SONAME_MAJOR 4 //!< Shared object library major version number.
#define NV_TENSORRT_SONAME_MINOR 1 //!< Shared object library minor version number.
#define NV_TENSORRT_SONAME_PATCH 3 //!< Shared object library patch version number.

in /usr/include/aarch64-linux-gnu/NvInfer.h it gives

attempting to open cache file ../../model/MobileNetSSD_deploy.caffemodel.1.tensorcache
loading network profile from cache...
createInference
2_softmax
2_detection_out
createInference_end
Bindings after deserializing:
Binding 0 (data): Input.
Binding 1 (detection_out): Output.
Allocate memory: input blob
allocate data
Allocate memory: output blob
allocate output
0.160521 , 0.179899 , 0.351957 , 0.714393
0.580017 , 0.174322 , 0.784316 , 0.736432
0.413608 , 0.159625 , 0.595889 , 0.726053
0.000298329 , 0.373827 , 0.209366 , 0.635824
0.338725 , 0.275225 , 0.457445 , 0.66792
0.236015 , 0.194621 , 0.419907 , 0.727357
0.303909 , 0.27831 , 0.374488 , 0.605555
0.155144 , 0.36166 , 0.227306 , 0.66111
0.51822 , 0.325927 , 0.622801 , 0.700739
0.394651 , 0.231804 , 0.47706 , 0.608612
0.137143 , 0.229672 , 0.221782 , 0.533472
0.805657 , 0.316405 , 0.892924 , 0.630012
0.300952 , 0.424454 , 0.492521 , 0.719962
0.286159 , 0.322753 , 0.40298 , 0.690346
0.0566538 , 0.329556 , 0.308375 , 0.661921
0.0507843 , 0.427002 , 0.283697 , 0.584319
0.0226981 , 0.503433 , 0.254028 , 0.661167
0.155144 , 0.36166 , 0.227306 , 0.66111
0.000288054 , 0.478121 , 0.384935 , 0.732708
0.177882 , 0.204922 , 0.833934 , 0.77984
0.00984597 , 0.476518 , 0.926414 , 0.993194
0 , 0.42333 , 0.284814 , 0.658839
0.00240755 , 0.413551 , 0.0775371 , 0.622686
0 , 0.485509 , 0.13807 , 0.610267
0.0330818 , 0.352652 , 0.187628 , 0.472965
0.00896727 , 0.417102 , 0.14612 , 0.57539
0.0558874 , 0.539785 , 0.231099 , 0.723213
0.3153 , 0.327529 , 0.464135 , 0.600311
0.101324 , 0.584882 , 0.267479 , 0.710831
0 , 0.242435 , 0.580386 , 0.747623
0.0305635 , 0.384605 , 0.13506 , 0.503647
0 , 0.524516 , 0.492444 , 0.857428
0.289528 , 0.388107 , 0.390684 , 0.52256
0.0543259 , 0.567055 , 0.154639 , 0.671858
0.0781092 , 0.595458 , 0.182098 , 0.661447
0.000298329 , 0.373827 , 0.209366 , 0.635824
0.177882 , 0.204922 , 0.833934 , 0.77984
0.0566538 , 0.329556 , 0.308375 , 0.661921
0.0286795 , 0.0125149 , 0.841668 , 0.597584
0 , 0.242435 , 0.580386 , 0.747623
0.115826 , 0 , 0.901469 , 0.271405
0 , 0.136121 , 0.474437 , 0.481822
0.00295338 , 0.38081 , 0.111221 , 0.538201
0 , 0 , 0.417294 , 0.323122
0.0330818 , 0.352652 , 0.187628 , 0.472965
0 , 0.216587 , 0.269543 , 0.494779
0.507684 , 0.53427 , 1 , 0.862502
0.333895 , 0 , 1 , 0.38035
0.455289 , 0.21072 , 1 , 0.762372
0 , 0.513718 , 0.761851 , 0.833565
9.11802e-05 , 0 , 0.391233 , 0.624349
0.047419 , 0.359526 , 0.135353 , 0.443856
0.0566538 , 0.329556 , 0.308375 , 0.661921
0.000298329 , 0.373827 , 0.209366 , 0.635824
0.420242 , 0.172266 , 0.582335 , 0.645928
0.0305635 , 0.384605 , 0.13506 , 0.503647
0.338725 , 0.275225 , 0.457445 , 0.66792
0 , 0.00798881 , 0.363116 , 0.738297
0 , 0 , 0.250309 , 0.486873
0 , 0.216587 , 0.269543 , 0.494779
0.12848 , 0.47811 , 0.543521 , 0.759041
0.00240755 , 0.413551 , 0.0775371 , 0.622686
0.476562 , 0.821771 , 1 , 0.971772
0.00896727 , 0.417102 , 0.14612 , 0.57539
0.000288054 , 0.478121 , 0.384935 , 0.732708
0.743723 , 0.351719 , 1 , 0.988543
0.287176 , 0.050057 , 0.701922 , 0.759556
0.155144 , 0.36166 , 0.227306 , 0.66111
0.361402 , 0.489218 , 0.773844 , 0.766053
0.047419 , 0.359526 , 0.135353 , 0.443856
0.568539 , 0.174267 , 0.809359 , 0.753875
0.139102 , 0 , 0.451539 , 0.535674
0.0126136 , 0.822241 , 0.551643 , 0.978369
0.764938 , 0 , 1 , 0.506245
0.143949 , 0.188334 , 0.381128 , 0.730597
0.64835 , 0 , 0.995357 , 0.754575
0 , 0.603846 , 0.388759 , 1
0.245007 , 0 , 0.745477 , 0.367092
0.0330818 , 0.352652 , 0.187628 , 0.472965
0.143508 , 0.704635 , 0.745253 , 0.987022
0.543932 , 0 , 0.858451 , 0.549616
0.0507843 , 0.427002 , 0.283697 , 0.584319
0.5334 , 0.485179 , 0.938008 , 0.780334

(process:11109): Gtk-WARNING **: Locale not supported by C library.
	Using the fallback 'C' locale.
distroyPlugin

nvidia@tegra-ubuntu:~/chinmay/MobileNet-SSD-TensorRT$ ./build/bin/mobileNet
attempting to open cache file ../../model/MobileNetSSD_deploy.caffemodel.1.tensorcache
cache file not found, profiling network model
../../model/MobileNetSSD_deploy_iplugin.prototxt
../../model/MobileNetSSD_deploy.caffemodel
CaffeParser: Could not open file ../../model/MobileNetSSD_deploy.caffemodel
CaffeParser: Could not parse model file
mobileNet: /home/nvidia/chinmay/MobileNet-SSD-TensorRT/tensorNet.cpp:105: bool TensorNet::caffeToTRTModel(const char*, const char*, const std::vector<std::__cxx11::basic_string >&, unsigned int, std::ostream&): Assertion `blobNameToTensor != nullptr' failed.
Aborted (core dumped)

I have downloaded the source code and trying to run it with the steps provided. When ./build/bin/mobileNet command used, it shows above error.

Package versions installed
tensorrt 4
cudnn 7
opencv 3

Can you help me with above issue?

I want to figure out how to fix it T.T

@Ghustwb
I found your question at NVIDIA's forum, here: https://devtalk.nvidia.com/default/topic/1043279/same-tensorrt-code-get-different-result/?offset=10#5307682
I refer to the same github code as you mentioned to build SSD tensorRT version https://github.com/saikumarGadde/tensorrt-ssd-easy
And now I came cross the same problem as you raised at nvidia's forum:

1. the result will change every time i run the inference, although the variation is slight, the result is below
   
   the results are defferent between two detectons about image-71.
2. there is big difference between the 1080ti and jetson tx2,
   
   the confidence about image-71 is much lower, I don't know why.

I tried your code, although your implementation is mobilenet-SSD, seems you have solved this problem? I run many times at 1080ti and they have the same coordinates value, and the result at jetson tx2 has slightly differenence, the results are close to each other.

Could you give me some advices? I would be very grateful.

I tried to use python api to quantize the SSD model.
I want to make use of plugin, so I copy several files and make a plugin.
Following the idea demonstrated by TensorRT official documentation, /usr/src/tensorrt/samples/python/fc_plugin_caffe_mnist
However, I encountered the problem while I want to import plugin, it shows fcplugin.so: undefined symbol: Z11cudaSoftmaxiiPfS.

Anyone has some ideas on what to do??

I found your code don not update about Readme.

Thanks for your great work, but I can only achieve 40+ fps, so I'm wondering how can I achieve 50+ fps. Furthermore, I want to ask that do you know how to implement fp16 inference with Iplugin layer???

hi Ghustwb,

for batch processing of images, i changed BATCH_SIZE = 9 and const size_t size = width * height * sizeof(float3) * BATCH_SIZE and filled the unifrom_data accordingly (uniform_data[volImg + line_offset + j] = ... where volImg changes according to batch index). I also changed first dim: 9 in .prototxt. To obtain detections i used indexing like output[ (batch_idx*100 + k) * 7 + 1 ]. There is only person (class 3) in my frames. For batch_id = 0 i got correct results (one person). But for other bathces detection data is like 1, 0.80123, 0.257755, 0.764545, 0.86875, 0.909765 # 2, 7.80123, 0.364645, 0.26875, 0.809765, 0.654343.

what i am doing wrong, can you help me? thanks!

In tensorRT 4.0 doc:
It supports softmax on C, H, W dimension

attempting to open cache file ../../model/MobileNetSSD_deploy.caffemodel.1.tensorcachecache file not found, profiling network model
../../model/MobileNetSSD_deploy_iplugin.prototxt
../../model/MobileNetSSD_deploy.caffemodel
mobileNet: /home/nvidia/MobileNet-SSD-TensorRT-master/pluginImplement.h:75: nvinfer1::Dims Reshape::getOutputDimensions(int, const nvinfer1::Dims*, int) [with int OutC = 5]: Assertion `(inputs[0].d[0])*(inputs[0].d[1]) % OutC == 0' failed.Aborted (core dumped)

Why am i getting this error?, Is there somewhere else than other than pluginImplement.h, and pluginImplement.cpp i need to make changes? Please kindly help

when i run mobileNet binary in release mode, it core dump, how can i reslove it?

I found a strange error
I have changed my prototxt and code as you did, , and my classification is 9
but when I use my caffemodel ,there is no result of output, I don't know why

By the way , when I use your model and protoxt , I comment the "void cudaSoftmax".
But the result is correct
If I don't do that , there is an error , "undefined reference to 'cudaSoftmax(int, int,float*,float*)'

Could you please give me some suggestions?Thanks

I met an error in TX2（TesnorRT 4）,Assertion 'blobNameToTensor != nullptr' failed.
Have you tried with TensorRT 4?

Hi @Ghustwb , Thanks for your works, i've been looking into this repo for quite a while. Thanks . Now I'm here with another simple doubt. Is it possible to do detection like this with real-time fps, with Yolov3(original version, not tiny-yolo) in Tensor-rt and jetson-tx2?

Hi @Ghustwb , I tested with your existing model in this repo, and its very accurate and almost perfect. Can you please tell me which are the classes in your existing model. i saw car and person and one must be background, but can you tell me which are the other classes?

I also would like to train a model as better and perfect as yours, can you tell me which dataset you used?, did you transfer learn or trained from scratch? and how many iterations did you go?

I've run the program successfully,but the display result is wrong, do you know why??

my version is 2.4.9, it could not been compiled successfully.

想要训练自己的模型，请问有训练模型的代码么？

thanks for your work and how to get layer output for debug with tensorrt?

@Ghustwb ,hello, i use tensorrt to improve my model ,but the result is same as before, do you know the reason?

Hi, Thanks for work. Very fast work. But after starting the program a lot of objects detection (demo data):
allocate output 0 , 0 , 0.126316 , 0.126316 0.00823669 , 0 , 0.149658 , 0.167737 0.0315789 , 0 , 0.231579 , 0.126316 0.1135 , 0 , 0.254921 , 0.167737 0.136842 , 0 , 0.336842 , 0.126316 0.218763 , 0 , 0.360184 , 0.167737 0.242105 , 0 , 0.442105 , 0.126316 0.324026 , 0 , 0.465447 , 0.167737 0.347368 , 0 , 0.547368 , 0.126316 0.429289 , 0 , 0.570711 , 0.167737 0.452632 , 0 , 0.652632 , 0.126316 0.534552 , 0 , 0.675974 , 0.167737 0.557895 , 0 , 0.757895 , 0.126316 0.639816 , 0 , 0.781237 , 0.167737 0.663158 , 0 , 0.863158 , 0.126316 0.745079 , 0 , 0.8865 , 0.167737 0.768421 , 0 , 0.968421 , 0.126316 0.850342 , 0 , 0.991763 , 0.167737 0.873684 , 0 , 1 , 0.126316 0 , 0.00823669 , 0.167737 , 0.149658 0.0427892 , 0.00823669 , 0.325632 , 0.149658 0.200684 , 0.00823669 , 0.483527 , 0.149658 0.358579 , 0.00823669 , 0.641421 , 0.149658 0.516473 , 0.00823669 , 0.799316 , 0.149658 0 , 0 , 0.126316 , 0.126316 0.00823669 , 0 , 0.149658 , 0.167737 0.0315789 , 0 , 0.231579 , 0.126316 0.1135 , 0 , 0.254921 , 0.167737 0.136842 , 0 , 0.336842 , 0.126316 0.218763 , 0 , 0.360184 , 0.167737 0.242105 , 0 , 0.442105 , 0.126316 0.324026 , 0 , 0.465447 , 0.167737 0.347368 , 0 , 0.547368 , 0.126316 0.429289 , 0 , 0.570711 , 0.167737 0.452632 , 0 , 0.652632 , 0.126316 0.534552 , 0 , 0.675974 , 0.167737 0.557895 , 0 , 0.757895 , 0.126316 0.639816 , 0 , 0.781237 , 0.167737 0.663158 , 0 , 0.863158 , 0.126316 0.745079 , 0 , 0.8865 , 0.167737 0.768421 , 0 , 0.968421 , 0.126316 0.850342 , 0 , 0.991763 , 0.167737 0.873684 , 0 , 1 , 0.126316 0 , 0.00823669 , 0.167737 , 0.149658 0.0427892 , 0.00823669 , 0.325632 , 0.149658 0.200684 , 0.00823669 , 0.483527 , 0.149658 0.358579 , 0.00823669 , 0.641421 , 0.149658 0.516473 , 0.00823669 , 0.799316 , 0.149658 0 , 0 , 0.126316 , 0.126316 0.00823669 , 0 , 0.149658 , 0.167737 0.0315789 , 0 , 0.231579 , 0.126316 0.1135 , 0 , 0.254921 , 0.167737 0.136842 , 0 , 0.336842 , 0.126316 0.218763 , 0 , 0.360184 , 0.167737 0.242105 , 0 , 0.442105 , 0.126316 0.324026 , 0 , 0.465447 , 0.167737 0.347368 , 0 , 0.547368 , 0.126316 0.429289 , 0 , 0.570711 , 0.167737 0.452632 , 0 , 0.652632 , 0.126316 0.534552 , 0 , 0.675974 , 0.167737 0.557895 , 0 , 0.757895 , 0.126316 0.639816 , 0 , 0.781237 , 0.167737 0.663158 , 0 , 0.863158 , 0.126316 0.745079 , 0 , 0.8865 , 0.167737 0.768421 , 0 , 0.968421 , 0.126316 0.850342 , 0 , 0.991763 , 0.167737 0.873684 , 0 , 1 , 0.126316 0 , 0.00823669 , 0.167737 , 0.149658 0.0427892 , 0.00823669 , 0.325632 , 0.149658 0.200684 , 0.00823669 , 0.483527 , 0.149658 0.358579 , 0.00823669 , 0.641421 , 0.149658 0.516473 , 0.00823669 , 0.799316 , 0.149658 0 , 0 , 0.126316 , 0.126316 0.00823669 , 0 , 0.149658 , 0.167737 0.0315789 , 0 , 0.231579 , 0.126316 0.1135 , 0 , 0.254921 , 0.167737 0.136842 , 0 , 0.336842 , 0.126316 0.218763 , 0 , 0.360184 , 0.167737 0.242105 , 0 , 0.442105 , 0.126316 0.324026 , 0 , 0.465447 , 0.167737 0.347368 , 0 , 0.547368 , 0.126316 0.429289 , 0 , 0.570711 , 0.167737 0.452632 , 0 , 0.652632 , 0.126316 0.534552 , 0 , 0.675974 , 0.167737 0.557895 , 0 , 0.757895 , 0.126316 0.639816 , 0 , 0.781237 , 0.167737 0.663158 , 0 , 0.863158 , 0.126316 0.745079 , 0 , 0.8865 , 0.167737 0.768421 , 0 , 0.968421 , 0.126316 0.850342 , 0 , 0.991763 , 0.167737 0.873684 , 0 , 1 , 0.126316 0 , 0.00823669 , 0.167737 , 0.149658 0.0427892 , 0.00823669 , 0.325632 , 0.149658 0.200684 , 0.00823669 , 0.483527 , 0.149658 0.358579 , 0.00823669 , 0.641421 , 0.149658 0.516473 , 0.00823669 , 0.799316 , 0.149658

when I run the command: ./mobileNet
then I get the following errors:
attempting to open cache file ../../model/MobileNetSSD_deploy.caffemodel.1.tensorcache loading network profile from cache... createInference The engine plan file is incompatible with this version of TensorRT, expecting 5.0.6.3got -2100382470.0.2.1852793695, please rebuild. createInference_end Bindings after deserializing: Segmentation fault (core dumped)
But the requirement lists shows TensorRT4 is OK.

It looks like I should install tensorRT 5 according to the error, but when I run dpkg -l | grep TensorRT
it shows:
...~/MobileNet-SSD-TensorRT$ dpkg -l | grep TensorRT ii graphsurgeon-tf 5.0.6-1+cuda10.0 arm64 GraphSurgeon for TensorRT package ii libnvinfer-dev 5.0.6-1+cuda10.0 arm64 TensorRT development libraries and headers ii libnvinfer-samples 5.0.6-1+cuda10.0 all TensorRT samples and documentation ii libnvinfer5 5.0.6-1+cuda10.0 arm64 TensorRT runtime libraries ii python-libnvinfer 5.0.6-1+cuda10.0 arm64 Python bindings for TensorRT ii python-libnvinfer-dev 5.0.6-1+cuda10.0 arm64 Python development package for TensorRT ii python3-libnvinfer 5.0.6-1+cuda10.0 arm64 Python 3 bindings for TensorRT ii python3-libnvinfer-dev 5.0.6-1+cuda10.0 arm64 Python 3 development package for TensorRT ii tensorrt 5.0.6.3-1+cuda10.0 arm64 Meta package of TensorRT ii uff-converter-tf 5.0.6-1+cuda10.0 arm64 UFF converter for TensorRT package

My TensorRT is 5!
Someone can help me out with it? Thanks.

Thanks @Ghustwb for this repo. Finally it worked. I trained a model in with 5 classes and it worked. But how to make it work for 7 classes? I mean where all we have to make changes?

现在我也碰到这问题，没找到答案。
Jetpack 3.3已经是最新的版本了

I have my own MobileNet caffe model, and I encountered with the problem of custom layer of DepthWise Convolution. I saw you change your MobileNet model's dw_conv into group_conv, so I mimic your move and run with TensorRT 7.2.

It works, however, it has outputs like Weights for layer conv2_1/dw doesn't exist.. Is this the problem of the model or the problem of this move? Have you met with the same problem?

Hope to hear from you, thanks.

compile success,but error on run time like below output.

attempting to open cache file /home/ubuntuuser/MobileNet-SSD-TensorRT/model/MobileNetSSD_deploy.caffemodel.1.tensorcache
loading network profile from cache...
createInference
The engine plan file is incompatible with this version of TensorRT, expecting 5.0.6.3got -2100382470.0.2.1852793695, please rebuild.
createInference_end
Bindings after deserializing:
Segmentation fault (core dumped)

I am trying to use my own trained SSD model with two classes (including background).
I have made changes to PluginImplement.h ,Pluginimplement.cpp and prototxt files as given here. (#9)
I deleted the cache file as it was showing engine not compatible error on Jetson Nano.

While running, I get the error :

Weights for conv0 doesn't exist
CaffeParser : ERROR : Attempting to access NULL weights
Parameter check failed at: ../builder/Network.cpp::addConvolution::66, condition : kernelWeights.values !=nullptr
error parsing layer type Convolution index 0

hi，你好，我从你的博客过来，我看到你的博客里说解决了检测抖动的问题，是softmax层的问题，可以请问一下是softmax哪里实现的问题吗？谢谢。

I ran the tensorrt code on Tesla V100 (TensorRT 5.1.2.2, CUDA10.1) and K40 (TensorRT 4.0.1.6, CUDA 8.0), and got different results on a few images.

Would you please give me some adivse ?

thanks in advance

Hi @Ghustwb , I've been looking into this repo, its an awesome work, i've been looking for the exact same. But I'm trying to train a caffe model with my custom dataset. Please help me

花了一天时间在ubuntu上，还是没编译通过

请问要用自己训练的模型，类别数不一样，需要修改哪些地方呢？另外您说用5类做训练和检测，这个5类包含背景么？

As the subject. I'm trying to build on Jetson TX2.

/home/nvidia/MobileNet-SSD-TensorRT/mathFunctions.h:8:19: fatal error: cblas.h: No such file or directory

I am trying to run the same code in the repository for the some time. However, the memory usage increases every second.

RAM 7756/7851MB (lfb 5x4MB) CPU [5%@2035,0%@2035,4%@2035,21%@2035,5%@2035,5%@2035] [email protected] [email protected] GPU@48C [email protected] Tboard@45C [email protected] PMIC@100C [email protected] VDD_IN 4667/9426 VDD_CPU 626/4204 VDD_GPU 184/841 VDD_SOC 884/1035 VDD_WIFI 0/10 VDD_DDR 1284/1893
RAM 7757/7851MB (lfb 5x4MB) CPU [5%@2035,0%@2035,0%@2035,11%@2035,14%@2035,6%@2035] [email protected] [email protected] GPU@48C [email protected] Tboard@45C [email protected] PMIC@100C [email protected] VDD_IN 4431/9423 VDD_CPU 627/4202 VDD_GPU 147/840 VDD_SOC 885/1035 VDD_WIFI 0/10 VDD_DDR 1284/1892
RAM 7758/7851MB (lfb 5x4MB) CPU [10%@2035,0%@2035,0%@2035,19%@2035,15%@2035,8%@2035] [email protected] [email protected] GPU@48C [email protected] Tboard@45C [email protected] PMIC@100C [email protected] VDD_IN 4519/9420 VDD_CPU 737/4200 VDD_GPU 184/840 VDD_SOC 885/1035 VDD_WIFI 0/10 VDD_DDR 1303/1892
RAM 7759/7851MB (lfb 5x4MB) CPU [12%@2035,0%@2035,0%@2035,17%@2035,4%@2035,4%@2035] [email protected] [email protected] GPU@48C [email protected] Tboard@45C [email protected] PMIC@100C [email protected] VDD_IN 4339/9416 VDD_CPU 627/4197 VDD_GPU 184/839 VDD_SOC 885/1035 VDD_WIFI 0/10 VDD_DDR 1284/1891
RAM 7759/7851MB (lfb 5x4MB) CPU [13%@2035,0%@2035,0%@2035,12%@2035,8%@2035,7%@2035] [email protected] [email protected] GPU@48C [email protected] Tboard@45C [email protected] PMIC@100C [email protected] VDD_IN 4394/9413 VDD_CPU 627/4195 VDD_GPU 147/839 VDD_SOC 885/1035 VDD_WIFI 0/10 VDD_DDR 1284/1891

And after some time the process is killed. Can you suggest a method to avoid the memory leakage?

#9
This issue seems like it only useful if the number of classes is less than 5.
I want to use the full 21 classes, so I applied app the changes on the link, but they didn't help.
Should I make a new caffemodel.1.tensorcache?
If I should, how can I make tensorcache?

graphics memory leak！！！
don't know where the problem is？

HI @Ghustwb ,i am new in tensorrt.
when i run make
i meet "pluginImplement.h:17:0: warning: "CHECK" redefined
#define CHECK(status) "
how can i do?

Hi, I use your code to do SSD-variant model conversion, It has the error:
NvPluginSSD.cu:428: virtual void nvinfer1::plugin::PriorBox::configure(const nvinfer1::Dims*, int, const nvinfer1::Dims*, int, int): Assertion H == inputDims[0].d[1] failed.

so I check all priorbox layer, the bottom, and top blobs are the same with origin prototxt, then I want to print each layer output shape, I check tensorNet.cpp TensorNet::getTensorDims(const char* name) function,

engine->getNbBindings() when this line of code is executed, the program has the segmentation fault.

Can you give me some advice? Thank you in advance!

thanks your work, I compile the project on pc and sucess, and I run "mobilenet" bin on tx2?

Hi!

When I use the demo to de video detection, It has the error:
runtime.cpp (16) - Cuda Error in allocate: 2 terminate called after throwing an instance of 'nvinfer1::CudaError'
Can you give me some advice?

I run your code in tx1, when run mobilenet.bin and find this error, how to debug this error