Skip to content

Latest commit

 

History

History
251 lines (147 loc) · 11.5 KB

active.md

File metadata and controls

251 lines (147 loc) · 11.5 KB

Our meetups are Wednesdays at 11 AM EDT and Fridays at 6 PM EDT and Sundays at 11 PM EDT.


Active Projects

Select an area of interest and choose a TO DO to contribute. Contact Loren for details.

Start by installing a couple of our Website repos locally

Google Data Commons Timelines

TO DO: Hosting DataCommons locally with Flask - Vishnupriya and our GDC team IN PROGRESS: Javascript Timelines from Google Data Commons API - Mehul, Aishwrya, Vishnupriya IN PROGRESS: Python pull from Google Data Commons API - Zihan, Bhavna

More Data Commons Visualization Projects Observable with Data Commons - Data Loaders How-To

RealityStream ML

Run Models Colab

TO DO: Send URL hash # parameters to our Run Models CoLab using a webhook on Google Cloud (ChatGPT) - Abhishek L

Anvil with our CoLabs: Anvil Extras and Anvil and AnvilScope CoLab using Plotly - Soham

Vindhya is joining us Oct 24

Everybody's Home (Langchain from Repos)

For Everybody's Home Page

TO DO: Javascript Langchain

Our Langchain Repo - Dhananjay and Pranathi
Apply langchain.js to any GitHub repo in our home repo folder using API. Add field for entering Github account and repo name.

LangChain.JS: Setup javascript for Retrieval Augmented Generation (RAG) using a static Javascript page with LangChain.js, an API for a supported LLM like OpenAI, and data from a Google Sheet or Github Repo. And with LangChain removed

API Storage storage for LLM APIs using javascript



IN PROGRESS: Python Langchain

Python Retrieval Augmented Generation (RAG) Conversational RAG for 10 LLMs - Pradeep and Pranoy

Pradeep: Pinecone on AWS free 2 GB max Pranoy: DocArray in memory Vecto Store (database)

Google Notebook LM

Retrieval-Interleaved Generation (RIG)
Using Google Data Commons DataGemma AI - Zihan found that a paid Google plan was needed to avoid storage/memory errors/timeouts.

Innovations in Water Purification - Hyper Desalination - Content prep for RAG

International Trade Flow

TO DO: Processing Exiobase in CoLabs, displaying with javascript

Contributors: Gary, Satya, Himanshu, Sahil

International Trade Flow SQL Data Prep - Exiobase Colab and charts

Our little trade flow Sankey - Big Sankey - Our Fork with python 3.10 and bug resolved

TO DO: Chord Chart Data Prep

TO DO: Python to pull Harmonized Code (HS) lookups into Supabase - Initial work by Wenjie (and Chen)

TO DO: Sankey Industry eChart - eCharts uses a common echarts.min.js file which we'll load in Feed Viewer

TO DO: Python - Finalize our All the Places data by State and Zip

US EPA State Impacts

TO DO: Pull into SQL DuckDB

Javascript updates for US EPA impact reports
React Team - Mosaic column checkboxes
React Team - Commodity Totals in Jobs Reports Impact Label Pipeline - Starting point for duplicating US EPA RStudio in python

Open Footprints YAML Display

Open Footprints - Product Impact API, Add javascript tp Feed View

Food Nutrition Labels - Shali and Wenwei (Stella)

BuildingTransparency and Open Footprint labels

Bhavna, Yash

OpenWebUI LLM Location Data

IN PROGRESS - Dhananjay

OpenWebUI - Reactivates Ollama for Sync, Docker and location addons

Open WebUI (Projects) with Python and Retrieval Augmented Generation (RAG) for "context window" recency

Earthscape NextJS Chatbot UI - React, Supabase and NextJS Hosting using GitHub Pages

Add localsite.js to OpenWebUI - Use our Building Branches Locally techniques within our Datascape fork

Team Tools

Feed Player React

ML with Python and Google Data Commons

Storyboard Generator

Images and Videos from .CSV prompts for Interactive presentation backgrounds, Request Visualization

Tabulator, Industry Timelines, SQLite in Browser

Moonshot Challenges

Our most challenging projects - Take the leap