Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/qengineering/jetson-nano-image

Jetson Nano image with deep learning frameworks
https://github.com/qengineering/jetson-nano-image

cuda deep-learning jetson-nano mnn ncnn opencv pytorch sd-card-image team-viewer tegra tensorflow torch torchvision

Last synced: 7 days ago
JSON representation

Jetson Nano image with deep learning frameworks

Host: GitHub
URL: https://github.com/qengineering/jetson-nano-image
Owner: Qengineering
License: bsd-3-clause
Created: 2021-06-25T10:29:17.000Z (over 3 years ago)
Default Branch: main
Last Pushed: 2024-11-20T11:01:23.000Z (2 months ago)
Last Synced: 2025-01-14T21:09:24.395Z (14 days ago)
Topics: cuda, deep-learning, jetson-nano, mnn, ncnn, opencv, pytorch, sd-card-image, team-viewer, tegra, tensorflow, torch, torchvision
Homepage: https://qengineering.eu/install-tensorflow-2.4.0-on-jetson-nano.html
Size: 64.5 KB
Stars: 126
Watchers: 10
Forks: 23
Open Issues: 19
Metadata Files:
- Readme: README.md
- License: LICENSE

Awesome Lists containing this project

README

        # Jetson Nano DNN image

![output image]( https://qengineering.eu/images/SDcard32GBJetson.webp )


## A Jetson Nano image with OpenCV, TensorFlow and Pytorch

[![License](https://img.shields.io/badge/License-BSD%203--Clause-blue.svg)](https://opensource.org/licenses/BSD-3-Clause)



------------

## Installation.

- Get a 32 GB (minimal) SD-card which will hold the image. 

- Download the `JetsonNano.img.xz` image (**7.2 GByte!**) from our [Sync](https://ln5.sync.com/dl/4bdb25370/xewd7xqz-6khtpcs2-w7zais3c-8uqi2rcj) site. 

- Flash the image on the SD card with the [Imager](https://www.raspberrypi.org/software/) or [balenaEtcher](https://www.balena.io/etcher/).

- According to [issue #17](https://github.com/Qengineering/Jetson-Nano-image/issues/17#) only flash the xz directly, not an unzipped img image.

- Insert the SD card in your Jetson Nano **4 GB RAM** and enjoy.

- Username: ***jetson***

- Password: ***jetson***

- JetsonNano.img.xz md5sum: 621F2E4E0B7775E5293E2186C96E91AA

### GDrive.

In some parts of the world, getting a good solid connection to Sync is difficult. That is why we've also provided a copy on [Google Drive](https://drive.google.com/file/d/1xSKJQX2uuLI-ewShU8LP8FROC6ozyiwo/view?usp=sharing). However, Google Drive limits the number of daily downloads. Please be considerate and use Google Drive only if necessary.

------------

## Tips.

* If you are in need of extra space, you can delete the opencv and the opencv_contrib folder from the SD card. They are no longer needed since all libraries are placed in the /usr/ directory.

* Use a tool like [GParted](https://gparted.org/) `sudo apt-get install gparted` to expand the image to larger SD cards. We recommend a minimum of 64 GB. Deep learning simply requires a lot of space.



------------

## Pre-installed frameworks.

- [JetPack](https://developer.nvidia.com/embedded/jetpack) 4.6.0

- [OpenCV](https://qengineering.eu/deep-learning-with-opencv-on-raspberry-pi-4.html) 4.5.3

- [TensorFLow](https://qengineering.eu/install-tensorflow-2.4.0-on-raspberry-64-os.html) 2.4.1

- [TensorFlow Addons](https://qengineering.eu/install-tensorflow-2.4.0-on-raspberry-64-os.html) 0.13.0-dev

- [Pytorch](https://qengineering.eu/install-pytorch-on-raspberry-pi-4.html) 1.8.1

- [TorchVision](https://qengineering.eu/install-pytorch-on-raspberry-pi-4.html) 0.9.1

- [LibTorch](https://qengineering.eu/install-pytorch-on-raspberry-pi-4.html) 1.8.1 

- [ncnn](https://qengineering.eu/install-ncnn-on-jetson-nano.html) 20210720

- [MNN](https://qengineering.eu/install-mnn-on-jetson-nano.html) 1.2.1

- [JTOP](https://github.com/rbonghi/jetson_stats) 3.1.1 

- [TeamViewer aarch64](https://www.teamviewer.com/en/download/linux/) 15.24.5

Tensorflow 2.5 and above require CUDA 11. CUDA version 11 cannot be installed on a Jetson Nano due to incompatibility between the GPU and low-level software at this time, hence Tensorflow 2.4.1. Only when NVIDIA releases a JetPack with CUDA 11 will we be able to upgrade Tensorflow.

![output image]( https://qengineering.eu/images/Software_Jetson.png )



![output image]( https://qengineering.eu/images/JTOP_jetson.png )

------------

## OpenCV + TensorFlow.

Importing both TensorFlow and OpenCV in Python can throw the error: _cannot allocate memory in static TLS block_.


This behaviour only occurs on an aarch64 system and is caused by the OpenMP memory requirements not being met.


For more information, see GitHub ticket [#14884](https://github.com/opencv/opencv/issues/14884).


![output image](https://qengineering.eu/images/SwapImportOpenCVJetson.webp)

There are a few solutions. The easiest is to import OpenCV at the beginning, as shown above.


The other is disabling OpenMP by setting the -DBUILD_OPENMP and -DWITH_OPENMP flags OFF.


Where possible, OpenCV will now use the default pthread or the TBB engine for parallelization.


We don't recommend it. Not all OpenCV algorithms automatically switch to pthread.


Our advice is to import OpenCV into Python first before anything else.


------------

[![paypal](https://qengineering.eu/images/TipJarSmall4.png)](https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=CPZTM5BB3FCYL)