How to train a YOLO-NAS Pose Estimation Model on a custom dataset step-by-step

You will also get access to all the technical courses inside the program, also the ones I plan to make in the future! Check out the technical courses below 👇
___________________________________________________________

In this video 📝 we are going to take a look at how you can train a YOLO-NAS pose estimation model on a custom dataset. I'll walk you through every single step in the training pipeline, all the way from setting up and annotating your dataset, how to export it and use it in a Google Colab notebook. Then we will set up our data loaders and some configurations before starting the training of the custom model. At the end, we are of course going to take a look at inference with our best trained model.

If you enjoyed this video, be sure to press the 👍 button so that I know what content you guys like to see.

___________________________________________________________

___________________________________________________________

___________________________________________________________

📞 Connect with Me:

___________________________________________________________

___________________________________________________________

📷 My camera calibration software, CharuCo Boards, and Checker boards

___________________________________________________________

Timestamps:
0:00 Intro
0:19 YOLO-NAS Pose Documentation
3:46 Annotate Dataset
9:19 Imports
10:23 Sneak-peak Inference
12:03 Setup Trainer
12:32 Download the Dataset
13:40 Configuration File Structure
16:11 COCO Dataset Structure
17:26 Utility functions
18:51 Dataloaders
20:29 Keypoint Transformation
20:53 Setup Transform
21:40 Train Dataloaders
22:09 Instantasing The Model
22:47 Training Parameters
24:14 Training The Model
25:48 Training Results
28:55 Inference Custom Model
31:10 Outro

Tags:
#YOLO-NAS-Pose #CustomPoseEstimation #supergradients #yolonas #poseestimation

[NicolaiAI]

Join My AI Career Program
Enroll in My School and Technical Courses

[KieranAustin-zr]

Watched your previous YOLO-NAS video and was wondering about custom datasets...this video was perfectly timed! Cheers Nicolai

[vcarvewood]

Thanks for the tutorial. The NAS pose is a way more complicated to train in comparison to yolov8 pose. I think we will stay with ultralytics for long time even if v8 is 2% lower than newest sota models. But convenience of v8 is priceless, to train, run, in conda prompt is a single line. Thanks for Glen and all the members of ultralytics that work on such a user friendly tool.

[rajmeetsingh]

hi, can you provide the updated link to Tiger Dataset and Config file. Not able to open the link. Thanks

[MurcieloGODA]

Hey great work! I have a question though, what if I have two classes that come under the supercategory? Both of them use 2 keypoints. In this case how am i supposed to edit the contents of 'PoseEstimationDataset'? I have noticed that its only using category 0 in the code you provided 'joints = Any help would be appreciated!

[gala]

Hello, thank you so much for the video.

I'm trying to run it with a custom dataset imported from roboflow and I keep getting 'nan' in the validation set.

I get it in all metrics except for AR and AP.

Has anyone encountered this error? What can I do?

[ahmedprog]

Hi, I can only split the data into training and testing in any version of Yolo؟

[benschi]

Does anybody get the StopIteration Error when executed this line trainer.train(model=yolo_nas_pose,
training_params=train_params,
train_loader=train_dataloader,
valid_loader=val_dataloader
)