Project information

• Category: Software Eng - Generative AI
• Done for: Internship
• Project date: 2023
• Project URL: (Not available as it is in a private repository)

This project focuses on creating a JupyterLab extension to aid artists in the generation of images using diffusion models by providing a version control of their outputs. Unfortunately, I'm not able to provide more images than the ones available but there is a description of the project:

The JupyterLab PromptAll extension integrates a text prompt and image management system into the JupyterLab environment, enhancing its functionality for users who work with text-image pairings. This extension allows users to easily load, display, and manage text prompts and their associated images within a dedicated panel. By installing the extension through a series of steps involving conda, JupyterLab, and NodeJS, users can activate the PromptAll panel via the command palette. Once activated, the panel provides options to load individual images or entire folders of images and prompts, displaying them in an organized manner for easy navigation and management. The JupyterLab PromptAll extension was developed using a combination of JavaScript, TypeScript, and JupyterLab's extension framework. The development environment requires NodeJS for building the extension package, and the Mammoth JS library is used to convert .docx text prompts into plain text. The core development code is written in TypeScript, specifically under the `index.ts` file in the `./src` directory. JupyterLab's pinned version of yarn, called `jlpm`, is used to build and manage the extension's dependencies. The extension's features include viewing, uploading, and removing text prompts and images, which are presented in a user-friendly interface. Additionally, developers have the flexibility to extend and customize the extension's capabilities by adding new features, modifying the user interface, and handling new file formats. The documentation also provides detailed instructions on setting up a development environment, building the extension, and contributing to its development. These guidelines ensure that users can effectively utilize and enhance the PromptAll extension to suit their specific needs within the JupyterLab platform.