From the course: Deep Learning and Computer Vision: Object Detection with PyTorch
Join today to access over 24,900 courses taught by industry experts.
Challenge: Implementing object detection in GitHub Codespaces using PyTorch - PyTorch Tutorial
From the course: Deep Learning and Computer Vision: Object Detection with PyTorch
Challenge: Implementing object detection in GitHub Codespaces using PyTorch
(bright music) - [Instructor] In this exercise, you'll build an object detection pipeline using PyTorch inside GitHub Codespaces. Your goal is to detect wheat heads in real images using the Global Wheat Head Detection Dataset. You'll use a pre-trained YOLOv5 model, fine-tune it, and visualize the output with bounding boxes. Here's how you can approach it. First, open a GitHub repository and launch a Codespace within a Python environment. Next, download the Global Wheat Head Dataset from Kaggle. Then load and preprocess the images, making sure they're ready for training. Now bring in YOLO using PyTorch Hub and fine-tune it on your dataset. Finally, trying to generate predictions and visualize bounding boxes over the wheat heads. This challenge should take around 15 to 20 minutes to complete. Ready to give it a try? Pause here, work through the steps, and then join me in the next video where I will show you how I solved it.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.