🎥 New Video Series: YOLOv5 Inference Step-by-Step 🚀

Hi YOLO Community! :wave:

About Me:
I’m a Deep Learning and Computer Vision enthusiast, self-taught in coding, and passionate about exploring the potential of AI.

:star2: I’m excited to share a personal project I’ve been working on: a video series that walks through YOLOv5 inference step by step, designed to share my journey and help others dive deep into this powerful algorithm and better understand its inner workings. :books:

A huge thank you to Glenn Jocher and the many GitHub contributors for building and maintaining this amazing algorithm. :raised_hands: This series wouldn’t have been possible without their work and dedication, as well as the great documentation available on the Ultralytics website.

The series consists of 56 videos (yes, it’s thorough!) covering everything from model setup to detailed inference and detection processes.

I’ve structured it with both beginners and intermediate users in mind, and there’s also a companion Colab notebook for hands-on practice. :computer: My goal was to make YOLOv5 more accessible and share some of the insights I’ve gained along the way.

Here’s the playlist and the Colab notebook :notebook_with_decorative_cover: companion for hands-on practice to run sections of code as you follow along with the videos

I hope these videos are helpful to you all, and I’d love to hear any thoughts or feedback from the community. :speech_balloon: Feel free to reach out with any questions or share your own experiences! Thank you to everyone here for making this such a supportive space to learn and grow.

Happy learning! :tada:

Marc Tornero

2 Likes

@marctornero this is really cool! I took a quick look at your video playlist, and it’s clear you have put a lot of work into this, so thank you!

A few things I think might be helpful for anyone finding this post:

  1. My (personal) recommendation would be to start using ultralytics as it’s much more Pythonic and provides an easy to use interface with a lot of documentation. YOLOv5 is still around and certainly usable, but isn’t going to see significant updates like ultralytics will going forward.
  2. The playlist is likely for those looking to dive in deeply to the code and might not be for anyone “just trying to make it work.” I’m certainly of the mindset that having a strong understanding is highly valuable and so I appreciate the depth of information in the videos.

Of course ultralytics is an evolution of the code in the YOLOv5 repository, so these videos could definitely be helpful to start understanding the code that later became ultralytics. I’m definitely going to watch some of the videos myself, might be able to learn something new!

1 Like

Hi @BurhanQ
Thanks so much for your kind words! It means a lot to hear that from someone at Ultralytics, especially knowing the level of dedication you all bring to advancing the YOLO family of algorithms.

Your recommendation to start using ultralytics is very helpful, and I completely agree. Understanding YOLOv5 can give learners a grounding in the principles and workings of the YOLO architecture, helping them better appreciate the enhancements in later versions. This series dives deep into this v5 architecture for this reason—helping users build that foundational understanding.

At the same time, I can definitely see how jumping directly into the ultralytics repo and focusing on later versions can be very beneficial for users interested in the latest advancements and ease of use, and could even enhance the learning journey further if that is their focus.

Thank you again for taking the time to look over the series, and I’d love to hear any feedback or ideas that come to mind as you explore the videos.

Marc

1 Like