Scarf analytics pixel

Nov 8, 2024

From Pixels to Insights: Seamlessly Extracting and Visualizing Table Data with Unstructured and Hex

Nina Lopatina + Tarun Narayanan

Table extraction

Have you ever wanted to take a scanned image of a table and automagically create whatever plots you'd like of the data? Try out this notebook where we combine table extraction via Unstructured with Hex's magic for processing and visualizing the data. Try this notebook on your own data by copying it to your Hex workspace, and pointing it to a URL with your table of choice. And stay tuned for updates to see a no-code version of this workflow with Unstructured Platform.

We start with this pdf of several tables, that we ingest with Unstructured Serverless API:

And with a few preprocessing steps and natural language commands via Hex Magic, we have turned it into interactive graphs:




Check out our notebook to try for yourself, or sign up for our Serverless API to extract tables_as_html from your own image files!


Keep Reading

Keep Reading

Recent Stories

Recent Stories

Jan 16, 2025

Enterprise RAG: Why Connectors Matter in Production Systems

Unstructured

RAG

Jan 16, 2025

Enterprise RAG: Why Connectors Matter in Production Systems

Unstructured

RAG

Jan 16, 2025

Enterprise RAG: Why Connectors Matter in Production Systems

Unstructured

RAG

Dec 29, 2024

Transform files in S3 to Pinecone with Unstructured Platform with no code

Nina Lopatina

Unstructured

Dec 29, 2024

Transform files in S3 to Pinecone with Unstructured Platform with no code

Nina Lopatina

Unstructured

Dec 29, 2024

Transform files in S3 to Pinecone with Unstructured Platform with no code

Nina Lopatina

Unstructured

Dec 18, 2024

Introducing Unstructured Platform API for Programmatic Data Transformation

Unstructured

Unstructured

Dec 18, 2024

Introducing Unstructured Platform API for Programmatic Data Transformation

Unstructured

Unstructured

Dec 18, 2024

Introducing Unstructured Platform API for Programmatic Data Transformation

Unstructured

Unstructured