Have you ever spent too long scrolling through your photos to find that one picture you know you took? In this liveProject, you’ll solve that problem—with the power of deep learning!
Your challenge is to create a working image retrieval system that can search images based on text queries: type in ‘dog’ and you’ll get pictures of dogs. You’ll combine cutting-edge natural language processing and computer vision techniques, exploit semantic word representation methods, and train a neural network on social media data using image captions as annotations. This liveProject will develop your skills with PyTorch, and be an impressive piece for your data science portfolio.
This project is designed for learning purposes and is not a complete, production-ready application or solution.
This liveProject is for accomplished (intermediate) Python programmers who have developed basic neural networks and are comfortable setting up a training environment for machine learning. To begin this liveProject you will need to be familiar with:
- Intermediate Python
- Beginner PyTorch
- Basics of machine learning
- Basics of computer vision
- Basics of natural language processing
- Basics of neural networks
- Setting up a neural network training environment
you will learn
In this liveProject, you’ll learn how to extend deep concepts to both natural language processing and computer vision. You’ll develop experience with both PyTorch and deep learning libraries, and extend them to achieve nonstandard tasks.
- Train a Word2Vec word representation method with custom data
- Customize and train a CNN with PyTorch
- Compute similarities between words and images given their embeddings