Git Product home page Git Product logo

llm-playground's Introduction

LLM Playground

Welcome to the LLM Playground repository! This collection is dedicated to showcasing the power of Large Language Models (LLMs) through Google Colab notebooks. Explore a range of projects related to LLMs, including but not limited to those utilizing langchain and similar technologies.

About LangChain

LangChain is a tool for developers that enables the creation of chains using large language models (LLMs). These chains can generate text based on provided prompts. LangChain offers versatility in constructing different types of chains and can be integrated with existing LangChain constructs. Each step in the chain requires specific prompts, and input data must be formatted appropriately for optimal outputs. LangChain is also useful for feeding large documents into LLMs.

  1. Components: LangChain offers modular abstractions for components necessary to work with language models. Whether you're using the entire framework or just these components, they are designed for ease of use.

  2. Use-Case Specific Chains: Chains represent specific configurations of components tailored to accomplish particular use cases. These chains provide a higher-level interface for developers, making it easy to get started with specific applications. Chains are also customizable to suit your project's unique needs.

For more detailed information, refer to the LangChain documentation.

Projects

Explore a diverse range of projects in this repository, from hosting notebooks as a Flask API to intriguing applications of LangChain. You're also encouraged to contribute your own innovative projects related to LLMs.

Getting Started

To get started with the LLM Playground and LangChain, follow these steps:

Prerequisites

  • Python (3.7 or higher) installed on your system.
  • A basic understanding of Python programming and Jupyter Notebooks.

Installation

For detailed information and documentation on how to install LangChain, please refer to the LangChain documentation.

  1. Clone this repository to your local machine:

    git clone https://github.com/TanmayDoesAI/LLM-Playground
    cd LLM-Playground // Navigate to the project directory
    pip install -r requirements.txt // Install the required dependencies

Running the Notebooks

Explore the projects in this repository by running the included Jupyter notebooks. Here's how:

  1. Launch Jupyter Notebook:
jupyter notebook
  1. Open the notebook of your choice from the notebooks directory.
  2. Follow the instructions within the notebook to run and experiment with the LLM-powered applications.

Examples and Tutorials

To help you navigate the projects and LangChain, we've provided a link to example notebooks and tutorials LangChain Tutorials. These resources offer step-by-step guidance and insights into working with language models.

Contributing

Thank you for considering contributing to the LLM Playground repository! Whether you're fixing a bug, enhancing features, or introducing new projects, your contributions are highly appreciated.

To get started with contributing, please read our Contribution Guidelines for detailed information on how to:

  • Submit issues
  • Propose new projects
  • Make pull requests
  • Report bugs
  • Suggest enhancements

We welcome contributions from individuals of all expertise levels. Let's build a welcoming community around LLM projects together!

Happy coding!

llm-playground's People

Contributors

dhavisco avatar frantic-rabbit avatar petsamuel avatar selamanse avatar tanmaydoesai avatar

Stargazers

 avatar  avatar

Watchers

 avatar

llm-playground's Issues

Create a Pull Request Template

Description

A standardized pull request template can help contributors provide essential information about their changes. Let's create a template that prompts for details like a brief description of the changes, any related issues, and any special considerations.

Tasks

  • Create a new file named PULL_REQUEST_TEMPLATE.md.
  • Populate the template with prompts for necessary information.
  • Ensure the template is clear and concise, without any confusing elements.

Additional Information

Feel free to refer to other repositories' Pull Request templates for inspiration and best practices. A recommended resource would be a look at the trending repositories

Improve the Contributing Guidelines

Description

Let's make the Contribution Guidelines more beginner-friendly by including relevant resources for beginners. This can include links to articles, tutorials, or videos that provide an introduction to working LangChain or similar libraries like Autogen.

Suggested tasks

  • Add a section for beginners with curated resources.
  • Provide links to tutorials or articles that can help beginners get started.
  • Ensure the guidelines are clear and easy to follow.
  • Feel free to have more improvements with your own accord that would help the new contributors

Notebook to Finetune the Mistral Model on Colab

Description

Create a Colab notebook that guides users through the process of finetuning the Mistral model. Include detailed instructions, sample code, and any necessary resources for finetuning.

Tasks

  • Provide properly documented instructions in the code for finetuning the Mistral model.
  • It would be recommended that we have a smaller size dataset for people who just want to test things out, and when people would be satisfied they could change the dataset!

Additional Information

Consider providing links to a few compatible datasets for finetuning, so that people could search for more as per their requirements

Notebook to Run the New Mistral Model on Colab

Description

Create a Colab notebook that demonstrates how to run the new Mistral model. Provide step-by-step instructions along with sample code to guide users through the process.

Tasks

  • Include instructions and sample code for running the Mistral model.
  • Make sure that the notebook is well-documented and easy to follow.

Additional Information

Feel free to leverage existing Colab notebooks for similar models as a reference.

Using Pinecone DB as a Vector Database

Description

Create an implementation to use the Pinecone database as the vector database to create embeddings of new documents, and store them, create a doc searcher, and use that to talk to documents.

Using LLMs to Summarize a YouTube Video

Description

Create a Colab notebook that showcases how to use Large Language Models to summarize content from YouTube videos. Use existing libraries to fetch the transcript of the video and then pass it to a large language model (of your choice) and use it to summarise the transcript.

A markdown to navigate through the notebooks

Once we get more notebooks, it would be better to have a markdown where the notebooks can be easily navigated through, ideally having a title, tags, and then open with colab button

Welcome to LLM Playground for Hacktoberfest 2023

Welcome to LLM Playground!

We're thrilled that you're interested in contributing to the LLM Playground repository. Your contributions, no matter the size, are highly valued.

How to Contribute

  • Explore Existing Issues: Take a moment to peruse the existing issues. If something grabs your interest, fantastic! If not, don't hesitate to suggest your own ideas.

  • Claiming an Issue: Spotted an issue you'd like to tackle? Leave a comment to let us know, and we'll assign it to you.

  • Creating a New Issue: Have a fresh idea or come across a bug? Don't be shy—open a new issue. We're eager to hear your suggestions!

Contribution Guidelines

For detailed information on submitting issues, proposing new projects, making pull requests, reporting bugs, and suggesting enhancements, please refer to our Contribution Guidelines.

Important Notes

  • Notebook Guidelines: Ensure that your notebooks are well-documented with proper markdown and comments. This improves accessibility and readability for all users.

  • Patience with Pull Requests: Your patience during our review process is appreciated. We're dedicated to upholding the quality of contributions.

Get Started!

Eager to dive in? Peruse the open issues or initiate your own. If it aligns with our repository's objectives, we'll gladly assign it to you.

Thank you for being a part of this community. Happy coding!

Create an Issues Template

Description

A standardized issues template can help streamline the process of reporting bugs or suggesting enhancements. Let's create a template that prompts for details like steps to reproduce, expected behavior, and actual behavior.

Tasks

  • Create a new file named ISSUE_TEMPLATE.md.
  • Populate the template with prompts for necessary information.
  • Ensure the template is clear and user-friendly.

Additional Information

Feel free to refer to other repositories' Issue Templates for inspiration and best practices. A recommended resource would be a look at the trending repositories

Improve the README

Description

The current README provides a brief overview of the repository. Let's enhance it to provide clearer instructions on how to get started, including setup instructions and any prerequisites. Additionally, let's include examples or tutorials to help users navigate through the projects.

Suggested tasks

  • Update the README.md file with more detailed instructions.
  • Include examples or tutorials for better user guidance.
  • Ensure clarity and correctness in the language used.
  • Feel free to have more improvements with your own accord that would improve the accesibility

Additional Information

Feel free to refer to other repositories' READMEs for inspiration and best practices. A recommended resource would be a look at the trending repositories

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.