Official YOLO v7 Custom Object Detection on Colab

This YOLO v7 custom object detection tutorial is focused on training the custom model on Google Colab. After training, you can run inferencing locally or on Colab. The method is generic enough to train all seven variations of official YOLO v7 models. In our previous tutorial, we trained the base YOLO v7 custom object detector, and in this tutorial, we train YOLOv7x to compare the performance of these two variants.

** Code and custom dataset is available for our Patreon Supporters**

Custom dataset development in YOLO v7 format:

Setting up YOLO v7 locally:
---------------------------------------------
► Time Stamps:
Introduction: (0:00)
Setting up Colab: (0:29)
Mounting Google Drive: (0:47)
Cloning YOLO v7 Repository: (1:20)
Creating Custom Dataset: (2:23)
Training Custom YOLO v7 Model: (4:34)
Inferencing Custom Objects: (7:55)
Inferencing Custom Objects Locally: (9:00)
---------------------------------------------
► Links:
---------------------------------------------
Want to discuss more?

#TheCodingBug
---------------------------------------------
► My Other Tutorials:
---------------------------------------------
---------------------------------------------
DISCLAIMER: Links included in this description might be affiliate links. If you purchase a product or service with the links that I provide I may receive a small commission. There is no additional charge to you!

[brunospfc]

Another great tutorial, this channel should have far more recognition, some other tutorials take more time not to work, and they miss stuff, this is compact and complete, thanks

[CodeHacker]

Hey thank you so much I used both methods from both videos but this one worked flawlessly

[_Fangirling]

Thank you so much. You are the life saver!!!

[IzeIzeBaby]

Thanks alot this worked great but i didn't have to update yaml on my colab. It just worked :D
Training takes a while but i'm used to worse since playing around with dreambooth / stable diffusion

[limtless]

Thank you so much for the great tutorials!
I wanted to know if we can detect vehicle speed and capture over speeding vehicles real time? and capture their license plate. I am assuming we have to use yolov7 with opencv, but again I have no idea if that is possible or how to do it.

[haticeozdemir]

Hello. It's a very good video, well done. While I was testing on the test photo, I encountered the error "/bin/bash: -c: line 0: syntax error near unexpected token `('"). What could be the reason. Do you know?

[TECHNEWSUNIVERSE]

AssertionError: train: No labels in data/train/labels.cache. Can not train without labels.

Can you please help me with this error

[fangirlpurpose]

Thank you for the tutorial. Can u help me why I have this problem?

AssertionError: train: No labels in data/train/human.cache. Can not train without labels

while i've uploaded the labels in data folder

[abrarluvrabit]

Can we also do active learning in our coding like out of 20 testing images my system detect 19 correct images but only 1 incorrect detection so how can i tell my system about this 1 incorrect image detection so next time it will not repeat the same mistake again. and how can i do that without training the entire model again?

[srinadhila]

Sir, what is the learning rate used in the training? thank you sir

[mangaenfrancais]

Hi, Thank you, is there any way to use .pt file with Opencv ?

[vyshakhcp]

hey, so can i detect many images at a time. you wrote code for each images you are going to detect.so if i have to detect many images at once, then do i have to write code for each one of them in google colab and anaconda prompt????

[prasannaravijr]

Hi can you tell how to show bounding box and its values for an output of an image

[JavedKhan-toez]

Hello sir please make a video about how we connect Yolo v7 models to the web .my FYP model is ready only connect web app is remaining.
Thanks

[genextv]

Is it possible to obtain a saliency map from this model?

[TheAbishanth]

I used 100 epochs for training. it took 3 stops to complete the training, but in the end only last.pt file is genrated and i couldnt see best.pt

[abrarluvrabit]

Can you please also make a video on a Multi-label object dection in a single bounding box. I have a dataset that involves images with bounding boxes which have more than 1 label.
For example, I have 5 classes (0-4), then an image might have a bounding box with label "2" and another might have a bounding box with two labels "2, 4" ie it belongs to both categories. so how can we show two labels or classes on a single bounding box? please make a video on this or atleast provide some solution in your ans here i will be very thankful to you.

[aryashinde]

Hello
I followed your tutorial but I'm getting an error which says 'no labels in data/train ' cannot train without labels. I have followed every step but it's not able find the labelled images

[Nouf_]

Hi, I have error
torch.cuda.OutOfMemoryError: CUDA out of memory
I reduced batch size to 16, 8, 4 and 2 . I got the same error .
can u help me plz!

[MdAlamgirHossain-ovud]

sir, could you please give me idea how to test multiple image from the test folder or path in yolov7 model by using colab?
already, i had followed this video and the code.