From the course: Hands-On AI: Computer Vision Projects with Ultralytics and OpenCV

Unlock this course with a free trial

Join today to access over 24,700 courses taught by industry experts.

How to benchmarks different models

How to benchmarks different models - OpenCV Tutorial

From the course: Hands-On AI: Computer Vision Projects with Ultralytics and OpenCV

Start my 1-month free trial Buy for my team

How to benchmarks different models

“

- [Instructor] Once your model is trained and validated, the next step is to evaluate its performance in various real world scenarios. In this video, we will explore the benchmark mode. It is very useful method to access the speed and accuracy of the model across a range of export formats, including torchscript, ONNX, OpenVINO, and many others. Let's get started. In line number one, we have imported the YOLO module from Analytics Python package. In line number three, we are initializing the YOLO model. All the resources used in this video are available in the course code subdirectory 0309. Here in line number three, we are initializing the best dash detect dot PT file, available inside the subdirectory 0309. It's the same model we have created in the video two of this chapter, with the help of object detection algorithm for detection of apples moving on conveyor belt. We will benchmark this model across different export formats. Let's go to line number six to benchmark specific model.…

Contents