Official YOLO v7 COMPLETE Object Detection Tutorial | Windows & Linux

[@anishantony7687]

successfully run everything but saved images and videos don't have any bounding boxes and confidence. can you help me with this?

[@TheCodingBug]

You might've missed something. Check pytorch version also. Because it's not possible to have empty bounding boxes from pretrained model.

[@tanasakchiewchankul9739]

I have same problem with you. 😢

[@TheCodingBug]

@@tanasakchiewchankul9739 use --device cpu

[@wei-yenchen7610]

I have the same problem . My gpu is 1650ti 4G (laptop : G14) I don't konw the reason that the result don't have any bounding boxes and confidence. TheCodingBug is right. Only run on CPU , the problem is solved. add those code to the file named "detect.py" ↓↓↓ torch.cuda.is_available = lambda : False device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')

[@pathfinder8399]

@@wei-yenchen7610 change the value to; half = False in line 31 of detect.py, it works fine

[@Wingly113]

straight to the point, thanks for the video!

[@isrardawar9333]

Every step are well explained

[@GuilhermePiresSilvadeAlmeida]

Hello. Congratulations on the video. I couldn't find the images and videos you used in the example. Could you tell me how to get it? Thank you for your attention. Thanks!

[@TheCodingBug]

The link is in the description of the video.

[@shubhamsongire6712]

Thank you sir it helped me alot to learn latest yolo v7 model

[@danielleivy8180]

Very well explained 👍🏾

[@miloszivkovic6118]

Very easy and I made it. But my cpu was dying 😂 , i followed evry step and is it possible that yolo is still using my CPU?

[@MdFaiyazAhmedRME]

Thanks for this video. I have the same problem like others but i want clarify one thing. Do in need to use only pre trained images or i can simply download any image and use it. in my case No bounding box or score

[@TheCodingBug]

You can use any image. Try horse image that is included with the repository

[@MdFaiyazAhmedRME]

@@TheCodingBug Thanks for the helps sir.

[@MdFaiyazAhmedRME]

@@TheCodingBug AssertionError: train: No labels in data/train/labels.cache. Can not train without labels. can you tell me about this error. i have save all dataset in YOLO format etc.

[@piyushshinde5351]

@ThecodingBug successfully run everything but saved images and videos don't have any bounding boxes and confidence. Help me with this

[@TheCodingBug]

Set half=False. It should solve the problem... Or use YOLOv11 or Yolov8.

[@avr4dev]

Why do you set up pytorch separately? It is in requirements.txt (on your video as well)

[@TheCodingBug]

That's without cuda support. That's why I install it separately with cuda.

[@aashishmalhotra]

discovered your channel today , and so far content is good. what to expect in future. Will we be doing any projects in yolo?

[@TheCodingBug]

Thanks for the appreciation. Yes. I am working on couple of projects with YOLOv7. Stay tuned!!

[@dwnim]

thank you for the tutorial, help me alot. i'm newbie in this python environment. i have run the script with webcam and it worked. the problem is, i dont know how to end the webcam process. could you please help me of how to stop the process ? thanks in advance

[@TheCodingBug]

Press q or ESC. Otherwise, end script with cntrl+c

[@andrepascoa8579]

Hello im new to ML, where do you train your models in this example?

[@TheCodingBug]

I don't. I use pretrained model. To train a custom model, watch this video kzbin.info/www/bejne/Y4K6qX1mn554j5o

[@ShlomoOrna]

Very helpful video. Thank you. Could you add some explanation on how to run YOLOv7 POSE Estimation ? In a response or does it need a separate video ?

[@tokamohsen6696]

I tried to run it on many images but the run folder gets empty folders and no images produced

[@himanthrayagada5453]

@TheCodingBug thank so much for this video i take the combination of knifes and guns dataset after training it detects knife as an airplane but i didn't use label called airplane plz could you help me

[@TheCodingBug]

Change labels in text file before training as shown here kzbin.info/www/bejne/Y4K6qX1mn554j5o Or just don't load labels from model and redefine them as per your requirements (i. e. replace labels variable with your own)

[@verifili]

please can you do end to end toturial about real time objects detection, I mean how implement real time API.

[@TheCodingBug]

Hi. If you mean you want to see the results in realtime instead of saving them to hard drive, here is how you can do it. python detect.py --weights yolov7.pt --conf 0.25 --img-size 640 --nosave --view-img --source street2.mp4

[@verifili]

@@TheCodingBug I mean when I build project unsing yolo and wanna push this project in website , so in this case I need to build an API , the question is how we can do this!

[@TheCodingBug]

@@verifili Understood. I will by making a video next month.

[@verifili]

@@TheCodingBug thnak you 😁😁

[@ucduyvo4552]

Thank you so much, but I have problem with the time when running the stream rtsp link, it just maintains a short time, about 3-5 minutes and then no signal, show black screen, how can I solve this problem to increase the time streaming, maintain about 1 day, 1 month,...

[@annwang5530]

Question: could I train with YOLO in Jetson NX to create a model for inference in Raspberry pi? (Sorry I'm newbie)

[@TheCodingBug]

Yes you can do it. Just convert the model to onnx.

[@annwang5530]

@@TheCodingBug thanks, probably I will be using online Google service instead of Jetson

[@Kishi1969]

Dear Sir, Thanks you , i got a lots of Video from the site you told me and everything worked well with images and videos changing with different weights but im having problem with web cam its given me errors

[@TheCodingBug]

What's the error?

[@Kishi1969]

@@TheCodingBug Exception in thread Thread-1: Traceback (most recent call last): File "C:\Users\MMU\anaconda3\lib\threading.py", line 973, in _bootstrap_inner self.run() File "C:\Users\MMU\anaconda3\lib\threading.py", line 910, in run self._target(*self._args, **self._kwargs) File "C:\Users\MMU\Desktop\SA\yolov7\yolov7\utils\datasets.py", line 318, in update time.sleep(1 / self.fps) # wait time ZeroDivisionError: float division by zero C:\Users\MMU\anaconda3\lib\site-packages\torch\functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at C:\actions-runner\_work\pytorch\pytorch\builder\windows\pytorch\aten\src\ATen ative\TensorShape.cpp:2228.) return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] Traceback (most recent call last): File "C:\Users\MMU\Desktop\SA\yolov7\yolov7\detect.py", line 184, in detect() File "C:\Users\MMU\Desktop\SA\yolov7\yolov7\detect.py", line 67, in detect for path, img, im0s, vid_cap in dataset: File "C:\Users\MMU\Desktop\SA\yolov7\yolov7\utils\datasets.py", line 327, in __next__ if cv2.waitKey(1) == ord('q'): # q to quit cv2.error: OpenCV(4.6.0) D:\a\opencv-python\opencv-python\opencv\modules\highgui\src\window.cpp:1333: error: (-2:Unspecified error) The function is not implemented. Rebuild the library with Windows, GTK+ 2.x or Cocoa support. If you are on Ubuntu or Debian, install libgtk2.0-dev and pkg-config, then re-run cmake or configure script in function 'cvWaitKey'

[@TheCodingBug]

@@Kishi1969 It seems opencv installation has some issue. You can reinstall opencv by command: pip install opencv-python==4.5.5.64

[@Kishi1969]

@@TheCodingBug I installed the it but same errors and the Webcam , just open instantly and vanished

[@Kishi1969]

@@TheCodingBug Errors " WARNING: Environment does not support cv2.imshow() or PIL Image.show() image displays OpenCV(4.6.0) D:\a\opencv-python\opencv-python\opencv\modules\highgui\src\window.cpp:1267: error: (-2:Unspecified error) The function is not implemented. Rebuild the library with Windows, GTK+ 2.x or Cocoa support. If you are on Ubuntu or Debian, install libgtk2.0-dev and pkg-config, then re-run cmake or configure script in function 'cvShowImage'"

[@fatguyinalittlevive1107]

Thanks for the easy to follow tutorial! Do you have a tutorial for training your own YOLO v7 model?

[@TheCodingBug]

Yes. Following is for training on local machine. You can also find training with colab on my channel. kzbin.info/www/bejne/Y4K6qX1mn554j5o

[@serious-rc933]

I get PackagesNotFoundError: The following packages are not available from current channels: - pyhton=3.9

[@msirac9405]

CUDA initialization: CUDA driver initialization failed, you might not have a CUDA gpu. I got this error how can i fix it

[@TheCodingBug]

If you have nvidia GPU, update the drivers.

[@msirac9405]

@@TheCodingBug thanks for helping i like your contents keep going

[@msirac9405]

@@alihan1287 OMG yes how did u find me?

[@jeronimoalester4338]

Good afternoon. I was wondering if it's possible to get the coordinates of bounding boxes from yolo v 7 test/detection?

[@VasiliiSilvestrov]

Thank you very mush! Very helpful video! I want to share with the auditory a couple things: firstly: instead of downloading you can you git clone command. It will save lots of time: you won't need to extract and copy files. And secondly, сould you add the used pictures and videos in the description, please.

[@pjhsiao99]

Great video.

[@opensourcethings]

I want .cfg file and .weights file so that i use it in my python open cv code. is it possible ??

[@xinbert4941]

Thanks you help me to install yolov7 and let me know where is the result route

[@vijayalakshmisomu5419]

Hi I am trying to test the custom data on pretrained weights by chainging the test path in coco.yaml but its not working

[@krisnarengga]

is there any C or C++ implementation for YoloV7 ?

[@TheCodingBug]

I am not sure. But I'll be making darknet based tutorial for YOLOv7 in future.

[@kin_1997]

good content, thank you for sharing.

[@hindi_recap1]

How can we do instance segmentation using yolov7

[@TheCodingBug]

kzbin.info/www/bejne/qqKTeHxqe8abjbc

[@hindi_recap1]

@@TheCodingBug thanks 👍

[@RSF02]

Thanks for the useful tutorial. But it cant run and giving me this error FileNotFoundError: [WinError 2] The system cannot find the file specified: 'runs\\detect\\exp2' can you help mee ?

[@punityadav4672]

how get distance of specific object

[@TheCodingBug]

Will upload specific tutorial for that.

[@punityadav4672]

Please upload as soon as possible

[@neil7356]

I get the error: detect.py: error: unrecognized arguments: --source 0 😥

[@salahalghyaline3329]

Many thanks

[@naveenpaliwal3884]

Thanks for the video ! It helps a lot. How we can convert entire project in exe file

[@brunospfc8511]

Great video, this channel helped me with easy instructions :D, would you do it on custom dataset? thanks

[@TheCodingBug]

I am glad you found it helpful. Yes I will be uploading a tutorial on training yolo v7 on custom dataset next week.

[@Kishi1969]

@@TheCodingBug please kindly do on a custom dataset that would be great, i tried but having problems about where to place my. Yaml file 🙏🙏🙏

[@MEATHEADBooYA]

@@TheCodingBug I am interested on where the weights go now with Yolo V7. A custom dataset tutorial will be great!

[@Kishi1969]

Mind blowing and very educative. Where do you get your video? 🙏

[@TheCodingBug]

It's good to hear that you found it helpful. The videos are from royalty free websites.

[@Kishi1969]

@@TheCodingBug Thanks, i will look for video of such 🙏

[@hocklintai3391]

I tried using the camera to detect object real time, but it ended up with AttributeError: 'NoneType' object has no attribute 'Shape'. Any ideas how to fix that?

[@TheCodingBug]

The script is unable to detect your webcam. Maybe you're giving wrong webcam number.

[@code786yt2]

how to run real time object detection to scan my desktop.

[@abrarluvrabit]

Hello Sir thanks for the awesome video, from where i can get this street2.mp4 video file?

[@huangyou535]

Help me a lot！Tks！

[@satyamohanty-be7vo]

how to run it on webcam?

[@TheCodingBug]

Use 0 for --source instead of video.

[@TejashKatuwal]

qt.qpa.xcb: could not connect to display qt.qpa.plugin: Could not load the Qt platform plugin "xcb" in "/usr/local/lib/python3.8/dist-packages/cv2/qt/plugins" even though it was found. This application failed to start because no Qt platform plugin could be initialized. Reinstalling the application may fix this problem. Available platform plugins are: xcb. webcam is not working on colab

[@im_insaf]

Does it work for mac m1?

[@TheCodingBug]

Yes if you install Anaconda for Mac.

[@nawthazin9327]

Could I run this code in raspberry pi4?

[@TheCodingBug]

Yes. But use tiny version.

[@emanuel22223]

Can YOLOV7 be used with Darknet?

[@TheCodingBug]

No.

[@weavy2468]

is it detected?

[@ptginnovators5688]

Exception in thread Thread-1: Traceback (most recent call last): File "C:\Users\MMU\anaconda3\lib\threading.py", line 973, in _bootstrap_inner self.run() File "C:\Users\MMU\anaconda3\lib\threading.py", line 910, in run self._target(*self._args, **self._kwargs) File "C:\Users\MMU\Desktop\SA\yolov7\yolov7\utils\datasets.py", line 318, in update time.sleep(1 / self.fps) # wait time ZeroDivisionError: float division by zero C:\Users\MMU\anaconda3\lib\site-packages\torch\functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at C:\actions-runner\_work\pytorch\pytorch\builder\windows\pytorch\aten\src\ATen ative\TensorShape.cpp:2228.) return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] Traceback (most recent call last): File "C:\Users\MMU\Desktop\SA\yolov7\yolov7\detect.py", line 184, in detect() File "C:\Users\MMU\Desktop\SA\yolov7\yolov7\detect.py", line 67, in detect for path, img, im0s, vid_cap in dataset: File "C:\Users\MMU\Desktop\SA\yolov7\yolov7\utils\datasets.py", line 327, in _next_ if cv2.waitKey(1) == ord('q'): # q to quit cv2.error: OpenCV(4.6.0) D:\a\opencv-python\opencv-python\opencv\modules\highgui\src\window.cpp:1333: error: (-2:Unspecified error) The function is not implemented. Rebuild the library with Windows, GTK+ 2.x or Cocoa support. If you are on Ubuntu or Debian, install libgtk2.0-dev and pkg-config, then re-run cmake or configure script in function 'cvWaitKey' @TheCodingBug- Hey , thank you so much for this nice tutorial,plz help me in the above mentioned error. Note-I have already installed the opencv version that you have suggested.

[@shahadathossain3468]

detect.py", line 117, in detect s += '%g %ss, ' % (n, names[int(c)]) # add to string ~~~~~^^^^^^^^ IndexError: list index out of range

[@rizvee3977]

Traceback (most recent call last): File "C:\Users\User\yolov7\utils\google_utils.py", line 26, in attempt_download assets = [x['name'] for x in response['assets']] # release assets KeyError: 'assets' During handling of the above exception, another exception occurred: Traceback (most recent call last): File "detect.py", line 196, in detect() File "detect.py", line 34, in detect model = attempt_load(weights, map_location=device) # load FP32 model File "C:\Users\User\yolov7\models\experimental.py", line 251, in attempt_load attempt_download(w) File "C:\Users\User\yolov7\utils\google_utils.py", line 31, in attempt_download tag = subprocess.check_output('git tag', shell=True).decode().split()[-1] IndexError: list index out of range I am getting this error. What should I do?

[@TheCodingBug]

You didn't download the yolov7 weights file and placed alongside the code. If you have, make sure your spellings for the file name is correct for --weights parameter.

[@bravinraamnaidu2905]

Namespace(weights=['yolov7.pt'], source='street2.mp4', img_size=640, conf_thres=0.4, iou_thres=0.45, device='', view_img=True, save_txt=False, save_conf=False, nosave=True, classes=None, agnostic_nms=False, augment=False, update=False, project='runs/detect', name='exp', exist_ok=False, no_trace=False) YOLOR 2023-1-2 torch 1.10.0 CUDA:0 (NVIDIA GeForce 940MX, 2047.875MB) 'git' is not recognized as an internal or external command, operable program or batch file. Traceback (most recent call last): File "C:\FYP\YOLOV7\utils\google_utils.py", line 26, in attempt_download assets = [x['name'] for x in response['assets']] # release assets KeyError: 'assets' During handling of the above exception, another exception occurred: Traceback (most recent call last): File "C:\FYP\YOLOV7\detect.py", line 196, in detect() File "C:\FYP\YOLOV7\detect.py", line 34, in detect model = attempt_load(weights, map_location=device) # load FP32 model File "C:\FYP\YOLOV7\models\experimental.py", line 251, in attempt_load attempt_download(w) File "C:\FYP\YOLOV7\utils\google_utils.py", line 31, in attempt_download tag = subprocess.check_output('git tag', shell=True).decode().split()[-1] File "C:\Users\ACER\anaconda3\envs\yolov7\lib\subprocess.py", line 424, in check_output return run(*popenargs, stdout=PIPE, timeout=timeout, check=True, File "C:\Users\ACER\anaconda3\envs\yolov7\lib\subprocess.py", line 528, in run raise CalledProcessError(retcode, process.args, subprocess.CalledProcessError: Command 'git tag' returned non-zero exit status 1. so why i got error

[@reyvincentlazarte4506]

we have the same problem asking if you solve the problem

[@KUMARIPRIYANKA-f1f]

how to calculate precision and recall in this project