Fine-tuning PaliGemma for Custom Object Detection
Offered By: Roboflow via YouTube
Course Description
Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore a comprehensive tutorial on fine-tuning Google's open-source Vision-Language Model, PaliGemma, for custom object detection tasks. Follow step-by-step instructions to modify Google's notebook and train PaliGemma on a handwritten digits and math operations dataset from RF100. Dive into the JSONL format, learn how to deploy the fine-tuned model for real-world inference, and discover PaliGemma's capabilities in image captioning, visual question answering, and object detection. Gain insights into overcoming limitations and important considerations when working with this powerful model. Access additional resources, including GitHub repositories, research papers, and community sessions to further enhance your understanding and application of PaliGemma.
Syllabus
- PaliGemma Capabilities
- Environment Setup
- Dataset Format
- Downloading Pre-trained Model
- Loading Dataset
- Training and Evaluating the Model
- Deploying the Model
- Important Considerations
- Outro
- Community Session June 6th, 2024 at 08:00 AM PST / 11:00 AM EST / PM CET: https://roboflow.stream
Taught by
Roboflow
Related Courses
LLaRA: Supercharging Robot Learning Data for Vision-Language PolicyLaunchpad via YouTube Fine-tuning Florence-2: Microsoft's Multimodal Model for Custom Object Detection
Roboflow via YouTube Florence-2: The Best Small Vision Language Model - Capabilities and Demo
Sam Witteveen via YouTube Vision Language Models and PDFs: What You See Is What You Search - Haystack EU 2024
OpenSource Connections via YouTube Mastering Google's PaliGemma VLM: Tips and Tricks for Success and Fine-Tuning
Sam Witteveen via YouTube