search

tsutof / tiny_yolov2_onnx_cam

Tiny YOLO v2 Inference Application with NVIDIA TensorRT
45 stars 15 forks source link

Error when running #2

Open FernandoMen opened 4 years ago

FernandoMen commented 4 years ago

Hello, I'm having a problem while trying to run the script. After doing all the installation steps and defining the jetson clocks, I get this error when running python3 tiny_yolov2_onnx_cam.py --camera 0 (im using an usb camera):

python3: engine.cpp:868: bool nvinfer1::rt::Engine::deserialize(const void, std::size_t, nvinfer1::IGpuAllocator&, nvinfer1::IPluginFactory): Assertion `size >= bsize && "Mismatch between allocated memory size and expected size of serialized engine."' failed.

What could be the problem? Thanks

tsutof commented 4 years ago

FernandoMen Thank you for your contact. Which version of JetPack are you using? I've confirmed that my application worked on Jetson Nano with JetPack 4.2.1 and 4.2.2 (SD card images). For Jetson Nano, the application assumes Linux swapfile which is already configured as the default in JetPack 4.2.1 and 4.2.2. (If you have not, please refer to this page in JetsonHacks). If you are trying to run the application on environments other than Jetson, your problem might be related with TensorRT version mismatch.