search

ultralytics / yolov5

YOLOv5 🚀 in PyTorch > ONNX > CoreML > TFLite
https://docs.ultralytics.com
GNU Affero General Public License v3.0
49.77k stars 16.12k forks source link

detect question #3254

Closed trra1988 closed 3 years ago

trra1988 commented 3 years ago

❔Question

Additional context

hello when I run detect.py the code didn't show any error and the result save in the file, but the picture in file don't have any bbox and class in picture, the run result is below, thanks

(base) lin@lin:~/yolov5-master$ python detect.py --source data/images --weights yolov5s.pt --conf 0.25 Namespace(agnostic_nms=False, augment=False, classes=None, conf_thres=0.25, device='', exist_ok=False, hide_conf=False, hide_labels=False, img_size=640, iou_thres=0.45, line_thickness=3, name='exp', nosave=False, project='runs/detect', save_conf=False, save_crop=False, save_txt=False, source='data/images', update=False, view_img=False, weights=['yolov5s.pt']) YOLOv5 🚀 2021-4-27 torch 1.8.1+cu111 CUDA:0 (GeForce GTX 1660 Ti, 5944.625MB)

Fusing layers... Model Summary: 224 layers, 7266973 parameters, 0 gradients, 17.0 GFLOPS image 1/2 /home/lin/yolov5-master/data/images/bus.jpg: 640x480 Done. (0.045s) image 2/2 /home/lin/yolov5-master/data/images/zidane.jpg: 384x640 Done. (0.041s) Results saved to runs/detect/exp3 Done. (0.138s)

github-actions[bot] commented 3 years ago

👋 Hello @trra1988, thank you for your interest in 🚀 YOLOv5! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://www.ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install run:

$ pip install -r requirements.txt

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

Status

CI CPU testing

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

trra1988 commented 3 years ago

Hello @trra1988, thank you for your interest in YOLOv5! Please visit our Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://www.ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install run:

$ pip install -r requirements.txt

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

Status

CI CPU testing

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

Duplicate of #

Hello @trra1988, thank you for your interest in YOLOv5! Please visit our Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://www.ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install run:

$ pip install -r requirements.txt

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

Status

CI CPU testing

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

hello, Thanks for reply and its help me, but I still can not detect the object, this is my train result and the detect picture confusion_matrix F1_curve labels labels_correlogram P_curve PR_curve R_curve results results.txt test_batch0_labels test_batch0_pred test_batch1_labels test_batch1_pred test_batch2_labels test_batch2_pred train_batch0 train_batch1 train_batch2 thanks for the help.

trra1988 commented 3 years ago

this is the result of detect bus

529035872 commented 3 years ago

I had the same problem

glenn-jocher commented 3 years ago

@529035872 @trra1988 hi thanks for the bug notice! There seems to be an issue with Windows and/or Conda environments that causes detect.py to not detect anything sometimes. I'm not sure of the cause or the solution, so all I can do is point you to one of our verified environments below, where everything will work correctly:

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

Status

CI CPU testing

If 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.

529035872 commented 3 years ago

@glenn-jocher thanks for your reply, I've solve this problem. In yolo.py, export = False # onnx export if export=True, no objects will be detected

glenn-jocher commented 3 years ago

@529035872 yolo.py has no export variable in the entire file. Your code is out of date. Update your code before anything else.

529035872 commented 3 years ago

@glenn-jocher ok, thanks.

trra1988 commented 3 years ago

@glenn-jocher thanks for help

trra1988 commented 3 years ago

@529035872 hello, I check my yolo.py, there do not have this line(export = False # onnx export), this cause of you change your own code right?

wudashuo commented 3 years ago

If you use Windows and there is no detection result at all when using detect.py, then you may meet an environmental problem. You should downgrade your CUDA to 10.2 and reinstall pytorch1.8.1+cu102. I tested CUDA11.1, 11.2, and 11.3, with PyTorch 1.8.1, none of these worked, only CUDA10.2 worked.

trra1988 commented 3 years ago

@wudashuo ok, thanks

trra1988 commented 3 years ago

@wudashuo hello, thanks for the reply last time, I change my cuda to the vision 11.0 and the code is runing on the ubuntu, but I still can not dectect the object, have you run on ubuntu before? thanks

wudashuo commented 3 years ago

@wudashuo hello, thanks for the reply last time, I change my cuda to the vision 11.0 and the code is runing on the ubuntu, but I still can not dectect the object, have you run on ubuntu before? thanks

I met several environmental problems on Windows, but I never met a problem on Ubuntu, no matter CUDA10.2, CUDA11.1, or CUDA11.2. Have you changed the codes of plotting?

trra1988 commented 3 years ago

@wudashuo hello, thanks for the reply last time, I change my cuda to the vision 11.0 and the code is runing on the ubuntu, but I still can not dectect the object, have you run on ubuntu before? thanks

I met several environmental problems on Windows, but I never met a problem on Ubuntu, no matter CUDA10.2, CUDA11.1, or CUDA11.2. Have you changed the codes of plotting?

No ,I did not, I just detect with pretrain model

trra1988 commented 3 years ago

@wudashuo The problem is solved, I change my system to ubuntu18.04 and change cuda to 10.2, The problem is cause by vision of cuda, Thanks

glenn-jocher commented 3 years ago

@wudashuo @trra1988 another user found that FP32 inference worked for them but not FP16 inference (the default in detect.py).

Can you see if the problem also happens on your system with YOLOv5 PyTorch Hub models? The Hub models are loaded as FP32 and use a different inference pathway (with AMP) which does not convert them to FP16. Try this line:

python hubconf.py

You should see

Fusing layers... 
Model Summary: 224 layers, 7266973 parameters, 0 gradients
Adding AutoShape... 
YOLOv5 🚀 v5.0-112-g9f3a388 torch 1.8.1 CPU

image 1/5: 720x1280 2 persons, 2 ties
image 2/5: 720x1280 2 persons, 2 ties
image 3/5: 1080x810 4 persons, 1 bus
image 4/5: 1080x810 4 persons, 1 bus
image 5/5: 320x640
Speed: 100.4ms pre-process, 249.1ms inference, 1.7ms NMS per image at shape (5, 3, 640, 640)
Saved zidane.jpg, zidane.jpg, image2.jpg, bus.jpg, image4.jpg to runs/hub/exp5
github-actions[bot] commented 3 years ago

👋 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 ⭐!