My video on @spacy_io layout is now out! This is probably my favorite update from @explosion_ai (and that's saying something!) This package makes it simple to do region detection, table detection, and OCR with just 1 line of Python. #MachineLearning

Going live now! Join me for the first pilot stream on YouTube.

This is such a cool idea! Get an inside look into the life and work of an open-source developer and chat about NLP and more. I'll probably hang out in the chat for a bit as well 💙

@pyconweb 📍 Details & Tickets: https://t.co/GbxjfMTi3e 📆 Talk:

Writing a new talk on "What the history of the web can teach us about the future of AI" 🔮 Wanted to do this for so long because I think there are some great lessons & analogies here. I'll be presenting it for my keynote at @PyConWeb in Berlin on Jan 25 – hope to see you there!

🚀 Excited to introduce our keynote speaker, @_inesmontani , who will open the conference day with her talk, "What can the history of the web teach us about the future of AI?" 💫 Delighted to welcome you to PyCon+Web 2025, Ines! Join us for a day featuring 15+ speakers across

Thank you @_inesmontani for building spaCy , it tells us that , NLP isn't only deep learning , transformers and LLMs. It's so useful for certain tasks all at one place , more imp - production friendly. spaCy have most things all at one place and reminds - not to use sword for a

If you're using Prodigy and want to beta test it, check out this post for more details:

New plugin coming to Prodigy soon: a visual data dashboard! 📊✨ Manage, view and filter annotations and access data analytics and progress all in a neat web app. (Link for beta testing in thread!)

👩‍🏫 12 conferences, 5 keynotes, 13 talks 🎙️ 4 podcast interviews, 3 panel discussions ✍️ 11 blog posts, 4 personal posts, 3 case studies 🔮 focus of my work and vision 🖤 personal, podcasts, TV, books, music and concerts https://t.co/LKJuQnONnY

Happy Birthday to me! 🥳 It feels like a good opportunity to look back at 2024, an eventful but also difficult year for me. So here's my personal review, including travel, talks, writing and various things I did and enjoyed.

More details and registration:

Looking forward to joining the panel on PyLadies entrepreneurs and career development at @PyLadiesCon on Saturday 💖 It's a fully online conference with many cool people and talks, and you can still register!

Including: ▪️ why PDFs are a bad "source of truth" ▪️ implementation with @spacy_io and Docling ▪️ working with tabular data ▪️ end-to-end information extraction workflow ▪️ data collection and training Read more:

New post: From PDFs to AI-ready structured data 📃✨ A deep dive into document processing, layout analysis and a modular workflow for building end-to-end document understanding and information extraction pipelines using PDFs, Word documents, scans and more.

Repo:

The nice thing about it is that you can control how the table should be represented in the document text, e.g. for further information extraction and to make it easier for the model to handle. That's what I want to play with next, including rephrasing tables in natural language.

Many of you have been asking about PDF table extraction and I finally got around to experimenting with it 👀 Here's tabular data converted with Docling + TableFormer, anchored within the document text and accessible as a pandas.DataFrame:

$ pip install spacy-layout Code, examples and documentation: https://t.co/zXZWSTjgF1

The first version of my spaCy + Docling integration is here: 📚 process PDFs, Word documents & more 📝 structured text-based output via @spacy_io's Doc 🏷 layout spans for sections, headings etc. 🔮 apply NLP pipelines to PDFs ✂️ chunk your data for RAG