Data Formulator - Microsoft Open Source Data Analytics Tool (With LLM - Gemini 2.5 Pro)
Data Formulator - Microsoft Open Source Data Analytics Tool (With LLM - Gemini 2.5 Pro)
https://linktr.ee/learnbydoingwithsteven
🚀 Unlock Rich Data Visualizations with AI! Explore Microsoft's Data Formulator 📊
Welcome back to Learn by Doing with Steven! In this video, we dive into Microsoft's exciting new Data Formulator tool – an application that leverages the power of Large Language Models (LLMs) like Google's Gemini to help you transform data and create insightful visualizations iteratively.
Join me as I walk you through the process from start to finish:
Installation: Setting up Data Formulator on your local machine using pip.
Configuration: Connecting Data Formulator to the powerful Gemini 2.5 Pro model using an API key from Google AI Studio.
Exploration: Loading sample datasets and exploring the intuitive interface.
AI-Powered Visualization: Using the "Auto" mode to generate charts from natural language prompts (like asking for trends or correlation heatmaps!).
Manual Chart Building: Diving deeper into creating specific charts like Scatter Plots, Box Plots, Bar Charts (with facets!), Heatmaps, and more.
Understanding the Process: Seeing how the tool generates Python code in the background to perform data transformations.
This tool offers a fascinating glimpse into how AI can assist with data exploration and visualization tasks. While it might not replace sophisticated software like Power BI or Excel for complex dashboards, it's incredibly useful for quickly generating visualizations and understanding data through a unique, AI-driven workflow.
Perfect for data analysts, developers, AI enthusiasts, and anyone looking to streamline their data visualization process!
🕒 TIMELINE:
0:00 - Introduction: What is Data Formulator?
0:18 - How it Works: Using LLMs for Visualization
0:39 - Getting Started: Following GitHub Instructions
1:00 - Installation via Pip in VS Code
1:11 - Starting the Data Formulator Application
1:30 - Configuring the Model: Selecting Gemini
1:37 - Setting up the Gemini API Key (Google AI Studio)
1:47 - Inputting the Gemini 2.5 Pro Model Name
2:05 - Loading Sample Data (Unemployment Across Industries)
2:22 - Interface Overview & Visualization Challenges Prompt
2:54 - Exploring Data Fields & Initial Data View
3:16 - Using "Auto" Mode with a Natural Language Prompt ("trend of unemployment data")
4:03 - Viewing the AI-Generated Transformation Code
4:07 - Manual Chart Building: Box Plot Example
4:29 - Manual Chart Building: Bar Chart Example (Average & Median)
4:59 - Using Facets in Bar Chart (by Year)
5:03 - Manual Chart Building: Heatmap Example (Correlation)
5:26 - Manual Chart Building: Custom Point Plot Example (Opacity & Size)
5:56 - Creating Multiple Plots & Using the View Grid
6:13 - Manual Chart Building: Grouped Bar Chart Example (by Industry/Series)
6:41 - Using "Auto" Mode for Correlation Heatmap Prompt
7:07 - Manual Chart Building: Dotted Line Chart Example (Median Count by Month/Year)
7:16 - Key Takeaways & Final Thoughts
7:27 - Thank You & Call to Action
🔗 Useful Links:
Data Formulator GitHub Repo: https://github.com/microsoft/data-for...
Google AI Studio (Get your Gemini API Key): https://aistudio.google.com/app/apikey
Gemini Models Documentation: https://ai.google.dev/gemini-api/docs...
👍 If you found this video helpful, please give it a like and share it with your friends!
💬 Leave a comment below with your thoughts or questions about Data Formulator.
✨ Don't forget to subscribe for more content on AI, data analysis, and learning by doing!
#DataFormulator #Microsoft #DataVisualization #AI #ArtificialIntelligence #Gemini #LLM #LargeLanguageModel #DataAnalysis #Python #LearnByDoing #GoogleAI #MicrosoftResearch