Inference detectnet

Hello,
Please, I am trying to inference with detecnet and my network retrained, I am receiving these errors.

`[image] loaded ‘data/images/humans_0.jpg’ (413x300, 3 channels)
[cuda] no kernel image is available for execution on the device (error 209) (hex 0xD1)
[cuda] /home/sylia/Desktop/jetson-inference/c/tensorConvert.cu:236
[cuda] no kernel image is available for execution on the device (error 209) (hex 0xD1)
[cuda] /home/sylia/Desktop/jetson-inference/c/detectNet.cpp:734
[TRT] detectNet::Detect() – cudaTensorNormMeanRGB() failed
[OpenGL] glDisplay – set the window size to 413x300
[OpenGL] creating 413x300 texture (GL_RGB8 format, 371700 bytes)
[cuda] registered openGL texture for interop access (413x300, GL_RGB8, 371700 bytes)
[image] saved ‘data/test/humans_0.jpg’ (413x300, 3 channels)

[TRT] ------------------------------------------------
[TRT] Timing Report models/Person/ssd-mobilenet.onnx
[TRT] ------------------------------------------------
[cuda] invalid resource handle (error 400) (hex 0x190)
[cuda] /home/sylia/Desktop/jetson-inference/build/x86_64/include/jetson-inference/tensorNet.h:685
[TRT] Pre-Process CPU 0.00000ms CUDA 0.00000ms
[TRT] Total CPU 0.00000ms CUDA 0.00000ms
[TRT] ------------------------------------------------

[TRT] note – when processing a single image, run ‘sudo jetson_clocks’ before
to disable DVFS for more accurate profiling/timing measurements

[image] loaded ‘data/images/humans_1.jpg’ (585x388, 3 channels)
[cuda] no kernel image is available for execution on the device (error 209) (hex 0xD1)
[cuda] /home/sylia/Desktop/jetson-inference/c/tensorConvert.cu:236
[cuda] no kernel image is available for execution on the device (error 209) (hex 0xD1)
[cuda] /home/sylia/Desktop/jetson-inference/c/detectNet.cpp:734
[TRT] detectNet::Detect() – cudaTensorNormMeanRGB() failed
[OpenGL] glDisplay – set the window size to 585x388
[OpenGL] creating 585x388 texture (GL_RGB8 format, 680940 bytes)
[cuda] registered openGL texture for interop access (585x388, GL_RGB8, 680940 bytes)
[image] saved ‘data/test/humans_1.jpg’ (585x388, 3 channels)

[TRT] ------------------------------------------------
[TRT] Timing Report models/Person/ssd-mobilenet.onnx
[TRT] ------------------------------------------------
[cuda] invalid resource handle (error 400) (hex 0x190)
[cuda] /home/sylia/Desktop/jetson-inference/build/x86_64/include/jetson-inference/tensorNet.h:685
[TRT] Pre-Process CPU 0.00000ms CUDA 0.00000ms
[TRT] Total CPU 0.00000ms CUDA 0.00000ms
[TRT] ------------------------------------------------

[image] loaded ‘data/images/humans_2.jpg’ (1024x669, 3 channels)
[cuda] no kernel image is available for execution on the device (error 209) (hex 0xD1)
[cuda] /home/sylia/Desktop/jetson-inference/c/tensorConvert.cu:236
[cuda] no kernel image is available for execution on the device (error 209) (hex 0xD1)
[cuda] /home/sylia/Desktop/jetson-inference/c/detectNet.cpp:734
[TRT] detectNet::Detect() – cudaTensorNormMeanRGB() failed
[OpenGL] glDisplay – set the window size to 1024x669
[OpenGL] creating 1024x669 texture (GL_RGB8 format, 2055168 bytes)
[cuda] registered openGL texture for interop access (1024x669, GL_RGB8, 2055168 bytes)
[image] saved ‘data/test/humans_2.jpg’ (1024x669, 3 channels)

[TRT] ------------------------------------------------
[TRT] Timing Report models/Person/ssd-mobilenet.onnx
[TRT] ------------------------------------------------
[cuda] invalid resource handle (error 400) (hex 0x190)
[cuda] /home/sylia/Desktop/jetson-inference/build/x86_64/include/jetson-inference/tensorNet.h:685
[TRT] Pre-Process CPU 0.00000ms CUDA 0.00000ms
[TRT] Total CPU 0.00000ms CUDA 0.00000ms
[TRT] ------------------------------------------------

[image] loaded ‘data/images/humans_3.jpg’ (1000x667, 3 channels)
[cuda] no kernel image is available for execution on the device (error 209) (hex 0xD1)
[cuda] /home/sylia/Desktop/jetson-inference/c/tensorConvert.cu:236
[cuda] no kernel image is available for execution on the device (error 209) (hex 0xD1)
[cuda] /home/sylia/Desktop/jetson-inference/c/detectNet.cpp:734
[TRT] detectNet::Detect() – cudaTensorNormMeanRGB() failed
[OpenGL] glDisplay – set the window size to 1000x667
[OpenGL] creating 1000x667 texture (GL_RGB8 format, 2001000 bytes)
[cuda] registered openGL texture for interop access (1000x667, GL_RGB8, 2001000 bytes)
[image] saved ‘data/test/humans_3.jpg’ (1000x667, 3 channels)

[TRT] ------------------------------------------------
[TRT] Timing Report models/Person/ssd-mobilenet.onnx
[TRT] ------------------------------------------------
[cuda] invalid resource handle (error 400) (hex 0x190)
[cuda] /home/sylia/Desktop/jetson-inference/build/x86_64/include/jetson-inference/tensorNet.h:685
[TRT] Pre-Process CPU 0.00000ms CUDA 0.00000ms
[TRT] Total CPU 0.00000ms CUDA 0.00000ms
[TRT] ------------------------------------------------

[image] loaded ‘data/images/humans_4.jpg’ (1000x667, 3 channels)
[cuda] no kernel image is available for execution on the device (error 209) (hex 0xD1)
[cuda] /home/sylia/Desktop/jetson-inference/c/tensorConvert.cu:236
[cuda] no kernel image is available for execution on the device (error 209) (hex 0xD1)
[cuda] /home/sylia/Desktop/jetson-inference/c/detectNet.cpp:734
[TRT] detectNet::Detect() – cudaTensorNormMeanRGB() failed
[image] saved ‘data/test/humans_4.jpg’ (1000x667, 3 channels)

[TRT] ------------------------------------------------
[TRT] Timing Report models/Person/ssd-mobilenet.onnx
[TRT] ------------------------------------------------
[cuda] invalid resource handle (error 400) (hex 0x190)
[cuda] /home/sylia/Desktop/jetson-inference/build/x86_64/include/jetson-inference/tensorNet.h:685
[TRT] Pre-Process CPU 0.00000ms CUDA 0.00000ms
[TRT] Total CPU 0.00000ms CUDA 0.00000ms
[TRT] ------------------------------------------------

[image] loaded ‘data/images/humans_5.jpg’ (1024x606, 3 channels)
[cuda] no kernel image is available for execution on the device (error 209) (hex 0xD1)
[cuda] /home/sylia/Desktop/jetson-inference/c/tensorConvert.cu:236
[cuda] no kernel image is available for execution on the device (error 209) (hex 0xD1)
[cuda] /home/sylia/Desktop/jetson-inference/c/detectNet.cpp:734
[TRT] detectNet::Detect() – cudaTensorNormMeanRGB() failed
[OpenGL] creating 1024x606 texture (GL_RGB8 format, 1861632 bytes)
[cuda] registered openGL texture for interop access (1024x606, GL_RGB8, 1861632 bytes)
[image] saved ‘data/test/humans_5.jpg’ (1024x606, 3 channels)

[TRT] ------------------------------------------------
[TRT] Timing Report models/Person/ssd-mobilenet.onnx
[TRT] ------------------------------------------------
[cuda] invalid resource handle (error 400) (hex 0x190)
[cuda] /home/sylia/Desktop/jetson-inference/build/x86_64/include/jetson-inference/tensorNet.h:685
[TRT] Pre-Process CPU 0.00000ms CUDA 0.00000ms
[TRT] Total CPU 0.00000ms CUDA 0.00000ms
[TRT] ------------------------------------------------

[image] imageLoader – End of Stream (EOS) has been reached, stream has been closed
detectnet: shutting down…
detectnet: shutdown complete.`

Suggestions please thank you

Hi @sylia, if you are running on a non-Jetson device, you need to add your GPU’s compute capability to CMakeLists.txt here, and rebuild/reinstall:

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.