How to export and optimize YOLO-NAS object detection model for real-time with ONNX and TensorRT

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 cover how to export and optimize YOLO-NAS models to ONNX and TensorRT for real-time performance. We will cover the whole pipeline step by step so you can follow along. The speedups we can get from optimizing our models are crazy without really loosing accuracy. It's a ton of information to ingest, but very important in the world of deep learning and computer vision. This is how good real-world applications and projects are made.

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
0:44 YOLO-NAS Documentation
3:59 Install Cuda & TensortRT
4:55 Optimize to ONNX
11:36 Results ONNX
14:33 Convert to TensorRT
21:45 TensorRT Results
22:51 Outro

tags:
#YOLO-NAS #ONNX #TensorRT #supergradients

[NicolaiAI]

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

[taufiqahmed]

Hi can you make a video on how to prune and quantize yolov8 model after training it on custom dataset?

[ceachibogdan]

Is there no way to export with dynamic batching?

[akashpatel]

Can you please create notebook or video explaning, how to run the YOLO NAS model in Jetson Nano. I am unable to setup the dependency in jetson that is required for infercing YOLO NAS model.
I am using Jetson Nano 2GB model and Jetpack of 4.6.1 version. Can you please specify clear instructions of version of each dependency that needs to be downloaded in Jetson Nano and also how to download in Jetson Nano. Please🙏🙏🙏🙏

[ConstantinCatargiu]

hi! i followed all your steps but in the end the script made only one prediction even in the image are more than one object. Can you help to fix this issue? Thanks and congrats for your work!
:D :D :D

[murtadanazar]

nice work 👏, if you can create a same video but using yolov8 this will be so amazing ❤

[lautarocastillo]

Hi! Excellent tutorial and very good explanation throughout. Thank you!
I have seen the video and followed your steps, but at the end when executing the infer_trt.py file I have an error that I have also seen in your console:
[TRT] [E] 1: Error Code 1: Myelin (Error 400 destroying stream...

In your console you can also see the same error but you have generated the result.jpg, in my case not. Do you have any idea how to fix it or what I can check?
Thanks!

[waleedzaghloul]

An AI Character gain Help From another AI model To prove He is Human💀, Thanks it was a nice work🎉