-
-
Notifications
You must be signed in to change notification settings - Fork 16k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
OpenBlas issue and GPU not working on jetson Nano #3603
Comments
@Ammad53 your code is out of date. To update:
|
@glenn-jocher this is the latest pull. Same issue. python3 detect.py --weights burn_deep.pt --source video6.mov Fusing layers... |
@Ammad53 it appears you may have environment problems. Please ensure you meet all dependency requirements if you are attempting to run YOLOv5 locally. If in doubt, create a new virtual Python 3.8 environment, clone the latest repo (code changes daily), and RequirementsPython 3.8 or later with all requirements.txt dependencies installed, including $ pip install -r requirements.txt EnvironmentsYOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):
StatusIf this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are passing. These tests evaluate proper operation of basic YOLOv5 functionality, including training (train.py), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu. |
Any tutorial, how to run Yolov5 on jetson nano from scratch? |
lasted pull |
@Ammad53 oh, I did not realize you were on jetson nano. It's a popular deployment target but we do not have an official consolidated tutorial for it yet. This is actually one of the reasons we've launched the EXPORT competition, to be able to create official export tutorials to the most popular destinations:
|
@glenn-jocher python jetsonInfo.py NVIDIA Jetson Nano (Developer Kit Version) L4T 32.4.4 [ JetPack 4.4.1 ] Ubuntu 18.04.5 LTS Kernel Version: 4.9.140-tegra CUDA 10.2.89 CUDA Architecture: 5.3 OpenCV version: 4.1.1 OpenCV Cuda: NO CUDNN: 8.0.0.180 TensorRT: 7.1.3.0 Vision Works: 1.6.0.501 VPI: 0.4.4 Vulcan: 1.2.70 |
Run on docker container too. Gpu not working. Any solution Downloading https://github.com/ultralytics/yolov5/releases/download/v5.0/yolov5s.pt to yolov5s.pt... Fusing layers... |
@Ammad53: |
👋 Hello, this issue has been automatically marked as stale because it has not had recent activity. Please note it will be closed if no further activity occurs. Access additional YOLOv5 🚀 resources:
Access additional Ultralytics ⚡ resources:
Feel free to inform us of any other issues you discover or feature requests that come to mind in the future. Pull Requests (PRs) are also always welcomed! Thank you for your contributions to YOLOv5 🚀 and Vision AI ⭐! |
Seems like Jetson Nano is not using GPU (I've checked out in jtop) while inferencing... What may go wrong @glenn-jocher ? TERMINAL: python3 detect.py --source 0 --view-img --weights yolov5s.engine --imgsz 320 --device 0 --class 0 --conf-thres 0.4 --half detect: weights=['yolov5s.engine'], source=0, data=data/coco128.yaml, imgsz=[320, 320], conf_thres=0.4, iou_thres=0.45, max_det=1000, device=0, view_img=True, save_txt=False, save_conf=False, save_crop=False, nosave=False, classes=[0], agnostic_nms=False, augment=False, visualize=False, update=False, project=runs/detect, name=exp, exist_ok=False, line_thickness=3, hide_labels=False, hide_conf=False, half=True, dnn=False Python 3.7.0 required by YOLOv5, but Python 3.6.9 is currently installed Loading yolov5s.engine for TensorRT inference... |
@PiotrG1996 the device listed on this line is the device used for inference:
Note your source is limited to 7.5 FPS |
I really don't understand what limits me from reaching a higher FPS rate. Based on the source below, it should be possible to utilize Jetson Nano performance much better. How do I fix this blocker, stopping me from better results? ALXMAMAEV - TensorRT Jetson Nano Benchmark
|
@PiotrG1996 I would raise your question directly on the 3rd party repo |
That's perhaps what I would do as well. I just couldn't figure out, where is the difference between camera frame rate and inference output? 1 frame can be processed in 0.03s? How to improve this camera frame rate based on the tensorRT yolv5s.engine running on Jetson Nano? (Default code implementation from Ultralytics repository)
( 1 / 0.130s = 7.69 FPS ) => maybe this limits me from getting better results? |
@PiotrG1996 If the NMS (non-maximum suppression) time limit is being exceeded, it may indicate a bottleneck in the post-processing step of the inference pipeline rather than the inference itself. This could be limiting the FPS results. I recommend checking the post-processing code and profiling the execution to identify any potential improvements. Additionally, optimizing the NMS algorithm or the TensorRT engine configuration may help improve the overall performance. |
❔Question
Facing these issues.
OpenBlas warning
GPU not working
Due to this, it is taking 3 sec for singe detection.
I have provided snippets of the my virtual env
Thanks for helping me out
Additional context
Virtual Env installation that i did
apt install python3.8 python3.8-venv python3-venv
python3.8 -m venv env
source env/bin/activate
python3 --version
(env) tttt@tttt-desktop:~/Desktop/detection_framework$ pip list
Package Version
absl-py 0.12.0
cachetools 4.2.2
certifi 2021.5.30
chardet 4.0.0
cycler 0.10.0
google-auth 1.31.0
google-auth-oauthlib 0.4.4
grpcio 1.38.0
idna 2.10
kiwisolver 1.3.1
Markdown 3.3.4
matplotlib 3.4.2
numpy 1.20.3
oauthlib 3.1.1
opencv-python 4.5.2.54
pandas 1.2.4
Pillow 8.2.0
pip 21.1.1
protobuf 3.17.3
pyasn1 0.4.8
pyasn1-modules 0.2.8
pyparsing 2.4.7
python-dateutil 2.8.1
pytz 2021.1
PyYAML 5.4.1
requests 2.25.1
requests-oauthlib 1.3.0
rsa 4.7.2
scipy 1.6.3
seaborn 0.11.1
setuptools 56.0.0
six 1.16.0
tensorboard 2.5.0
tensorboard-data-server 0.6.1
tensorboard-plugin-wit 1.8.0
thop 0.0.31.post2005241907
torch 1.8.1
torchvision 0.9.1
tqdm 4.61.1
typing-extensions 3.10.0.0
urllib3 1.26.5
Werkzeug 2.0.1
wheel 0.36.2
Running Cpu
python3 detection_framework/detect.py --weights burn_best.pt --source videos/video6.mov
Using torch 1.8.1 CPU
Fusing layers...
OpenBLAS Warning : Detect OpenMP Loop and this application may hang. Please rebuild the library with USE_OPENMP=1 option.
OpenBLAS Warning : Detect OpenMP Loop and this application may hang. Please rebuild the library with USE_OPENMP=1 option.
OpenBLAS Warning : Detect OpenMP Loop and this application may hang. Please rebuild the library with USE_OPENMP=1 option.
Model Summary: 232 layers, 7246518 parameters, 0 gradients, 16.8 GFLOPS
OpenBLAS Warning : Detect OpenMP Loop and this application may hang. Please rebuild the library with USE_OPENMP=1 option.
OpenBLAS Warning : Detect OpenMP Loop and this application may hang. Please rebuild the library with USE_OPENMP=1 option.
video 1/1 (1/500) /home/tttt/Desktop/detection_framework/videos/video6.mov: 640x640 1 Burns, Done. (2.894s)
(env) tttt@tttt-desktop:~/Desktop/detection_framework$ python3 detection_framework/detect.py --weights burn_best.pt --source videos/video6.mov --device 0
Traceback (most recent call last):
File "detection_framework/detect.py", line 215, in
detect()
File "detection_framework/detect.py", line 40, in detect
device = select_device(opt.device)
File "/home/tttt/Desktop/detection_framework/detection_framework/utils/torch_utils.py", line 47, in select_device
assert torch.cuda.is_available(), 'CUDA unavailable, invalid device %s requested' % device # check availablity
AssertionError: CUDA unavailable, invalid device 0 requested
(env) tttt@tttt-desktop:~/Desktop/detection_framework$ nvcc -V
bash: nvcc: command not found
Similar issues
#474
#2863
The text was updated successfully, but these errors were encountered: