https://github.com/mongodb-developer/image-search-vector-demo

A Jupyter Notebook demonstrating how to use a multi-modal embedding model to build an image search engine.
https://github.com/mongodb-developer/image-search-vector-demo

ai jupyter-notebook python search vector

Last synced: about 17 hours ago
JSON representation

A Jupyter Notebook demonstrating how to use a multi-modal embedding model to build an image search engine.

• Host: GitHub
• URL: https://github.com/mongodb-developer/image-search-vector-demo
• Owner: mongodb-developer
• License: apache-2.0
• Created: 2024-01-29T09:12:41.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2024-01-29T09:18:08.000Z (11 months ago)
• Last Synced: 2024-11-06T02:26:26.960Z (about 2 months ago)
• Topics: ai, jupyter-notebook, python, search, vector
• Language: Jupyter Notebook
• Homepage:
• Size: 3.61 MB
• Stars: 10
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

README

        
# Image Search with MongoDB Atlas Vector Search

This repository contains a Jupyter Notebook demonstrating how to generate

vector embeddings for both text and images using a multi-modal embedding model.

## Getting Ready To Run The Notebook

The first thing you'll want to do is create a virtual environment using your favorite technique. I tend to use [venv](https://docs.python.org/3/library/venv.html), which comes with Python.

Once you've done that, install dependencies with:

```

pip install -r requirements.txt

```

You'll need to set an environment variable, `MONGODB_URI`, containing the connection string for your MongoDB cluster.

One more thing you'll need is an "images" directory, containing some images to index! I downloaded  [Kaggle's ImageNet 1000 (mini) dataset](https://www.kaggle.com/datasets/ifigotin/imagenetmini-1000), which contains lots of images at around 4GB, but you can use a different dataset if you prefer. The notebook searches the "images" directory recursively, so you don't need to have everything at the top-level.

Then you can fire up the notebook with:

```

jupyter notebook "Image Search.ipynb"

```

At the end of the tutorial, you'll be able to search for images with snippets

of text, like this:

![A screenshot of a line of code searching for "corgi in the snow" followed by a grid of animal photos.](readme_images/screenshot.png)