L-5 YOLOv5 on Jetson Nano | PyTorch & TorchVision Installation on Jetson Nano

our jetpack version is 5.1.1.. we used 1.14 pytorch verson. python version is 3.7.1.. but is shows ERROR: torch-1.14.9-cp36-cp36m-linux_aarch64.whl is not a supported wheel on this platform. what can i do for this?

@Aarohi , I am facing issues installing ultralytics on my jetson nano. As Ultralytivs wont support python below 3.7.0 and jetson nano runs 3.6.9. Any solutions?

any one tell me how to install cupy in this board with python3.6 ubuntu18.04

Mam can you please make a video on using yolov8 with csi camera jetson nano

How did you install ultralytics in the requirements.txt? I keep looking around and see that it's only available in python>=3.7

Im getting cannot open shared object file no such file or directory error when i import torch please help. My torch vision v0.11.1

hllo,
i got some errors while running the last command
RuntimeError: Running cythonize failed!

can u please help me with this error...

Hello Ma'am,
Thank you so much for your content!! It helped a lot, I learned so much , and you safed my group work!

HI there, Im trying to install yolov5 but now it downloads with a requirements.txt file that requires ultralytics package for yolov5 and python 3.6.9 is not compatible, I cant find a repo with tht requirements.txt file that you have

Thank you for the video! I have to ask when I run the csi-camera it shows late responses about a second difference, is that supposed to happen? Thank you!

Thank you so much for great video. May I know did you run Jupiter notebook in Virtual environment?

Here is my feedback about your videos: The videos are clear, go step by step, with good (not perfect) documentation. It is also easy to understand what you say, and your explanations are very clear. I followed your guidelines but it fails at the last stage because of versions mismatch. I think the root cause is that you had your SW installed already, so you don't face exactly what the user is facing. I would recommend with such a video series that you will go exactly as the user goes. Nothing pre-installed. Also, from time to time you missed packages that were needed, from same reason. I am aborting here because of that. Again, thank you for your investment.

Would a model trained with Yolov8 work with yoloV5?

When installing the requirement.txt, I get "ERROR: No matching distribution found for gitpython>=3.1.30". Is that a showstopper?

I am getting error while installing ultralytics-yolov5. Error message - "No matching distribution found for ultralytics-yolov5

Hi Arohi!
Could you provide your email id.
I got an error and want to resolve that issue by providing you errors screenshots.

Why you use jupyter instead of jetson terminal for downloading all the commands?

hello, good afternoon, could you please send me the script to use the CSI camera :)?

Hi, is it possible to run yolo on the jetson nano, and use stereo cameras so that you can get depth information? Can the jetson do both? Thank you for any insight

Hi, I'm facing a very annoying problem everytime i install torch vision, cause by the time i install it do it successfully but when i import it i get error in dependencies although i ensure to have them

Hello ma'am. Thank you so much for making this video and sharing your knowledge worldwidely. But Could I ask you something ?? I followed all of your videos realated this Jetson Nano. But I can't do $ pip3 install -r requirements.txt. It says ERROR : No matching distribution found for gitpython>=3.1.30 .My python version, torch, torchvision, and anything else's versions are exactly the same with you, but it seems like everyone excepts me can install requirements.txt . ( I edited requirements file as you told in this video). I would appreciate if you let me know what's the problem . Thank you so much

Hello Arohi,

I followed all the videos and the installations were successful. But I am facing an error wrt torchvision.
the installation is successful but it throws some error wrt the pillow module. the error is as below. Could you please suggest me. How can I fix this error.

File "<stdin>", line 1, in <module>
File "/home/innoright/.local/lib/python3.6/site-packages/torchvision-0.9.0-py3.6-linux-aarch64.egg/torchvision/__init__.py", line 7, in <module>
from torchvision import datasets
File "/home/innoright/.local/lib/python3.6/site-packages/torchvision-0.9.0-py3.6-linux-aarch64.egg/torchvision/datasets/__init__.py", line 1, in <module>
from .lsun import LSUN, LSUNClass
File "/home/innoright/.local/lib/python3.6/site-packages/torchvision-0.9.0-py3.6-linux-aarch64.egg/torchvision/datasets/lsun.py", line 2, in <module>
from PIL import Image
File "<frozen importlib._bootstrap>", line 971, in _find_and_load
File "<frozen importlib._bootstrap>", line 955, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 656, in _load_unlocked
File "<frozen importlib._bootstrap>", line 626, in _load_backward_compatible
File "/home/innoright/.local/lib/python3.6/site-packages/Pillow-9.5.0-py3.6-linux-aarch64.egg/PIL/Image.py", line 59, in <module>
File "<frozen importlib._bootstrap>", line 971, in _find_and_load
File "<frozen importlib._bootstrap>", line 951, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 894, in _find_spec
File "<frozen importlib._bootstrap_external>", line 1157, in find_spec
File "<frozen importlib._bootstrap_external>", line 1131, in _get_spec
File "<frozen importlib._bootstrap_external>", line 1112, in _legacy_get_spec
File "<frozen importlib._bootstrap>", line 441, in spec_from_loader
File "<frozen importlib._bootstrap_external>", line 544, in spec_from_file_location
File "/home/innoright/.local/lib/python3.6/site-packages/Pillow-9.5.0-py3.6-linux-aarch64.egg/PIL/_deprecate.py", line 1
SyntaxError: future feature annotations is not defined

hello all, is the YOLO v5 the latest version of YOLO that is compatible with the jetson nano? or could i use YOLO v8?

why the csi camera is faster than the rtsp camera ?

Hello, a problem has occurred where the terminal is not opening and there is an error message saying "Error occurred while checking for updates" at the top of the OS. Is there a solution for this?

mam i am getting a unable to establish ssl connection error
while installing pytorch

Hello, your version jetpack is 5.1.1 and torch-1.12.0-cp38-cp38-manylinux2014_aarch64.whl, torchvision-0.13.0-cp38-cp38-manylinux2014_aarch64.whl, version cuda 11.3
after install successfully torch, I print
>> import torch
>> torch.cuda.is_available()
>> false
please help me????

You should not run yolo as it is. Best option is to optimize it and build its tensorRT engine. I can see you are hardly getting 2-3FPS, with tensorRT, you can achieve 12-14FPS on Jetson Nano

Thank you very much, ma'am. This video helped me a lot.🤩

Hello ma'am, could u make a video for tensorflow installation as well...

Hello, we are big fans also... We are getting an error when trying to install pip3 install numpy torch-1.8.0-cp36-cp36m-linux_aarch64.whl the error returned is: "ERROR: torch-1.8.9-cp36-cp36m-linux_aarch64.whl is not a supported wheel on this platform." We have done lots of digging around and theorise it might have something to do with our Python version being 3.8 and so set up a virtual environment with Python 3.6 however this is still not doing it. Once we run the same install command as above the verbose message is in red and much longer and says there is a memory error? Can you help us please?

Hello, thank you so much for the lessons, I have tried many times to work on jetson nano and this is the first tutorial that works perfectly for me. Now, I have a question, (I'm quite noob on this), for running yolov5 on jetson nano you show us a jupyter notebook, you run it inside the environment, i mean you install jupyter notebook and run it from the terminal? and also Could you share us the jupyter notebook?

Thank you so much in advance!

Hello Arohi,,, Thank you very much for your effort
I have this problem have you encountered before
It appeared after writing this instruction as in the previous video (sudo apt-get -y upgrade)
------------------------------------the error--------------------------------------------------
update-alternatives: warning: skip creation of /usr/share/man/man1/mpiexec.1.gz because associated file /usr/share/man/man1/mpiexec.openmpi.1.gz (of link group mpirun) doesn't exist
Processing triggers for man-db (2.8.3-2ubuntu0.1) ...
Processing triggers for libc-bin (2.27-3ubuntu1.6) ...
Errors were encountered while processing:
nvidia-l4t-bootloader
nvidia-l4t-xusb-firmware
nvidia-l4t-initrd
E: Sub-process /usr/bin/dpkg returned an error code (1)
------------------------------------------------------------------------------------------------
Should the system be re-installed?

Great explanation,Could you provide a video how to implement yolov5 with deepsort and tensorRT in jetson nano

that's great is worked excellent, you can implement yolov5 with tensorRT and deepsort

I am big fan of you. you taught me how to train a dataset and now yolov5 Jetson deployment. Thanks for your instructive videos ;))

Thanks for all the videos are so useful, do you know if there are a way to accelerate and improve the fps on the jetson nano? I tried a lot of ways but a cant get more than 7 fps using de pretrained yolov5n model :(

Big follower of Ur presentation ,Thank you Aarohi..When i am using RTSP stream in yolov7 it freezes in its frame itself mam, any suggestions for that.

Hi there. Can you please import torch library and confirm what torch.cuda.is_available() returns? Maybe we can fix the subpar FPS on your video inference. Thanks in advance.

Hello lady Aarohi, all your videos are great, I am following your tutorials on the jetson nano, everything is working fine for me so far; I wanted to ask you if you can make a video on how to install MediaPipe on the jetson nano with OpenCV Cuda please.

Why yolov5, why not yolov7 or v8. In our organization jetson Orin board is purchased, that service man also installed yolov5. Is there any specific reason

Thank you