Weights & Biases

The repl uses Langchain and OpenAI's GPT-3 to create a Q&A bot for Weights & Biases Documentation. Use natural language queries to interact with docs.wandb.ai

Or fork it and swap in your our text data to give additional knowledge to GPT-3!

The script uses FAISS so turn on the Replit GPU instance for this script to run blazingly fast.

This script leverages the OpenAI GPT3 and Langchain to create a Q&A bot for the Weights & Biases documentation. Run the repl demo with your Openai API key to interactively query the documentation.

Created with ❤️ by Weights & Biases Sign up to Weights & Biases here to get started!

Weights & Biases is a machine learning experiment tracking, model checkpointing and data visualisation tool used by over 200,000 ML practitioners across the world.

You needs to turn on the Replit GPU instance for this script to train

This script leverages the computer vision components in the hugging face ecosystem to train a computer vision model to classify different types of food. You can swap out the dataset and model easily to customise it and make it your own!

Created with ❤️ by Weights & Biases Sign up to Weights & Biases here to get started!

🌟 Learn to train ML models in the Replit x Weights & Biases Hackathon from February 4th - 11th 🌟

Fork this repl and run it to get started with basic machine learning with Weights & Biases

Weights & Biases is a machine learning experiment tracking, model checkpointing and data visualisation tool used by over 200,000 ML practitioners across the world.

Running this repl will::

1. Run some dummy logging to show the basics of logging to Weights & Biases

2. Run a mini-experiment varying the amount of Dropout in your model and logging the results to Weights & Biases

To log results to your own Weights & Biases account, create an account at the link below, then enter your API key when prompted. Otherwise just select the option to log anonymously, without entering an API key.

Sign up to Weights & Biases here to get started!

🌟 To learn more about how to train your own ML models in Replit, sign up to the Replit x Weights & Biases Hackathon from February 4th - 11th 🌟

GPU Powered - When your fork this repl make sure to turn on the GPU option in your repl settings

Weights & Biases is a machine learning experiment tracking, model checkpointing and data visualisation tool used by over 200,000 ML practitioners across the world.

This repl will fine-tune a GPT2 model on Shakespeare using Andrej Karpathy's nanoGPT project

Sign up to Weights & Biases here to get started!

Click on "Open website" to view the app, it might take 1-2 minutes for the page to load. You should be able to see the app here

With this Photo Prediction App you can load a computer vision model into your repl.

Then a Gradio app will start where you can get predictions from it! ML on Replit has never been easier!

• Created with ❤️ by Weights & Biases