unstructured.io

Did you know you can batch process as many documents as you want with Unstructured Serverless API? We support over 25 different file types, and best of all, ingesting documents from a source is the fastest way to transform your data! Here's how to get started: 1. Watch this short Quickstart video: https://lnkd.in/gvg4-5x8 2. Grab your API Key: app.unstructured.io 3. Use this code sample?with your API key: https://bit.ly/3yC6LCB Don't forget: You get 1000/pages a day for FREE for the first 14 days! #WhateverItIsWeCanStructureIt

?? How do you remove Personally Identifiable Information (PII) from unstructured data used for RAG? ??In this new notebook, we walk through data preprocessing steps that ingest and transform unstructured documents, chunk them, remove PII using GLiNER, and then generate embeddings to be used for semantic search in RAG. ?? You can use this approach with Word documents, PDFs, emails, markdown, and many more unstructured data types. Notebook: https://lnkd.in/gt8P3G9j

?? Destination connector update! Unstructured now integrates with Milvus by Zilliz! With the new destination connector in the `unstructured-ingest` library, you can easily ingest data from 20+ sources, preprocess 25+ unstructured file types using the Unstructured Serverless API, chunk, embed, and seamlessly upload RAG-ready documents into Milvus's cloud-native & open-source vector database. With the below code snippet, you can batch process local unstructured files, chunk, embed them and load into https://lnkd.in/gcvnzjGc with just a few lines of code. Learn more in the docs: https://lnkd.in/g8Fqc3dD

?? Recently published documentation alert ?: Our SFTP?source?and?destination?connector documentation now includes a how-to video for setup source: https://lnkd.in/d25dxNAj destination: https://lnkd.in/d7zydqKK

?? Integration highlight! Looking for an easy way to integrate unstructured documents with LangChain? Try `UnstructuredLoader` with Unstructured Serverless API to process over 25 file types: pip install -qU langchain-unstructured unstructured-client

If you are building multi-agent RAG applications and need to extract data for your application, check out this tutorial for Unstructured x Langflow:

Check out Unstructured in the Microsoft commercial marketplace, and read our documentation for Unstructured API on Azure here: https://lnkd.in/gvTZJh2q

??Unstructured now supports embedding models from Mixedbread. Build ETL pipelines easily: * Seamlessly ingest unstructured docs from various sources * Process using Unstructured's Serverless API & Ingest lib * Add an embedding step with Mixedbread AI model to your pipeline via `EmbedderConfig`. Learn more about Unstructured Ingest in our documentation: https://lnkd.in/ei4c76x3

????♂? Looking for a fun weekend project to try out? Try this Scalable PDF Processing project with ?????????????????? and unstructured.io! In this project created by Tibor Mach you'll learn how to: - Extract and parse text from documents - Create vector embeddings for downstream tasks - Scale up document processing effortlessly - Version and persist datasets for reproducibility The best part? You can accomplish all this in less than 70 lines of code! Perfect for data scientists and ML engineers working with unstructured data. Key tools: - Unstructured.io for document processing - DataChain for scalable data handling and versioning Give it a try and unlock insights from your document collections! ???? Link to the tutorial in the comments! #DataScience #MachineLearning #WeekendProject #PDFProcessing

?? Ready to move past basic RAG? Join Unstructured and @MongoDB for a webinar: ?? Building Advanced RAG Apps with MongoDB, Unstructured, and LangGraph ?? September 26, 2024 | 12 PM ET What you’ll learn: ?? How to process raw documents using Unstructured’s Serverless API for chunking, embedding, and metadata extraction ?? Combine that metadata with MongoDB Atlas Vector Search for more relevant, efficient retrieval ??Use LangGraph to orchestrate RAG systems with a self-querying retriever Speakers: Apoorva Joshi, MongoDB Maria Khalusova, unstructured.io Walk away with actionable techniques you can apply right away: https://lnkd.in/eADVpDX9