lwc:hardware:jetson

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision		 Previous revision
lwc:hardware:jetson [2025/03/04 19:23] – [Installing pyTorch (WIP. May not be correct)] John Harrisonlwc:hardware:jetson [2025/03/05 17:12] (current) – [misc] John Harrison
Line 19: Line 19:
     * at that point I switched to ethernet as both units were connected to my router. I'm not sure if sticking with usb might have worked or would have been faster     * at that point I switched to ethernet as both units were connected to my router. I'm not sure if sticking with usb might have worked or would have been faster
  
-==== Installing pyTorch (WIP. May not be correct) ====+==== Installing pyTorch ==== 
 +//corrected from: [[https://docs.nvidia.com/deeplearning/frameworks/install-pytorch-jetson-platform/index.html]])//
   * install ''jetpack''. (This is not the same as installing the OS for Jetson): ''sudo apt install nvidia-jetpack''   * install ''jetpack''. (This is not the same as installing the OS for Jetson): ''sudo apt install nvidia-jetpack''
   * get version of CUDA: ''nvcc --version''   * get version of CUDA: ''nvcc --version''
-  * it's unclear if the following is needed (run as root) (from: [[https://docs.nvidia.com/deeplearning/frameworks/install-pytorch-jetson-platform/index.html]])+  * install ''cusparselt'':
 <code> <code>
-wget  +wget raw.githubusercontent.com/pytorch/pytorch/5c6af2b583709f6176898c017424dc9981023c28/.ci/docker/common/install_cusparselt.sh  
-raw.githubusercontent.com/pytorch/pytorch/5c6af2b583709f6176898c017424dc9981023c28/.ci/docker/ +sudo -s # set CUDA version and install as root 
-common/install_cusparselt.sh  +export CUDA_VERSION=12.1 # as an example. Use the version returned from ''nvcc --version''
-export CUDA_VERSION=12.1 # as an example. (example putting in the correct CUDA version caused this to fail. Only 12.1 worked)+
 bash ./install_cusparselt.sh bash ./install_cusparselt.sh
 </code> </code>
-  * install correct numpy version: ''sudo pip install numpy=1.26.1''+  * install correct numpy version: ''sudo pip install numpy==1.26.1''
   * download correct version of pytorch for jetpack version. Search for it at [[https://developer.download.nvidia.cn/compute/redist/jp/]]   * download correct version of pytorch for jetpack version. Search for it at [[https://developer.download.nvidia.cn/compute/redist/jp/]]
 +<code>
 +sudo pip3 install --no-cache http://<URL TO PYTORCH WHL>
 +</code>
 +
 +==== misc ====
 +  * [[https://elinux.org/Jetson_Zoo#ONNX_Runtime|onnxruntime for Jetson]]
 +  * if ''onnxruntime'' fails with message ''Invalid argument. Specify the number of threads explicitly'' fix with one-time cli command: ''sudo nvpmodel -m 0'' which forces the computer to with no throttling
 +  * ''tegrastats'' might show load on GPU?
 +    * ''tegrastats | cut -f 14 -d " "'' shows % load on GPU
 +  * you can get rid of the warning ''System Throttled due to Over-current'' by editing ''/usr/share/nvpmodel_indicator/nvpmodel_indicator.py''
  • lwc/hardware/jetson.1741137780.txt.gz
  • Last modified: 2025/03/04 19:23
  • by John Harrison