search

dusty-nv / jetson-inference

Hello AI World guide to deploying deep-learning inference networks and deep vision primitives with TensorRT and NVIDIA Jetson.
https://developer.nvidia.com/embedded/twodaystoademo
MIT License
7.78k stars 2.98k forks source link

Building the Project from Source Cmake Issue #758

Closed ianrheewhi closed 1 year ago

ianrheewhi commented 4 years ago

So I've been working on getting through the setup, but I I'm building the project from source rather than using the Docker. When I try to install PyTorch or cmake ../, I get the following issue:

CMake Error: The following variables are used in this project, but they are set to NOTFOUND. Please set them or make sure they are set and tested correctly in the CMake files: CUDA_CUDART_LIBRARY (ADVANCED) linked by target "jetson-inference" in directory /home/jsh/jetson-inference repeating message CUDA_TOOLKIT_INCLUDE (ADVANCED) used as include directory in directory /home/jsh/jetson-inference repeating message

-- Configuring incomplete, errors occurred! See also "/home/jsh/jetson-inference/build/CMakeFiles/CMakeOutput.log".

I've deleted the project folder and tried to do it again, but the issue persists. I've followed the instructions to a T, I don't know what I'm doing wrong; I've been trying to debug for hours. I would appreciate any and all input.

dusty-nv commented 3 years ago

Hi @ianrheewhi , can you check that you have the CUDA toolkit installed on your Jetson?

You should be able to find it under /usr/local/cuda

If you don't have CUDA, you would need to install it. I'm not sure which Jetson you are using or version of JetPack, but if you are on JetPack 4.3 or newer you should be able to run sudo apt-get install nvidia-jetpack. Otherwise, you could use SDK Manager tool to install it or if you are on Nano/Xavier-NX you can flash the NVIDIA SD card image.

DennisFaucher commented 3 years ago

@dusty-nv thanks for this. I just re-imaged my Xavier with the SDK Manager and nvidia-jetpack was not installed. I will install and see if my cmake errors go away. (Wow, download.nvidia.com is slow today. 47.4 kB/s 13h 42min 26s) "sudo apt-get install nvidia-jetpack" fixed the issue! Yay.