So I appear in the Celebration of 100 years of Bell Labs video with our LLM robots (a.k.a. IndustrialGPT) for less than a second :-D
https://www.youtube.com/watch?v=Fu_IundP5Ic&t=26s

New blog post: Understanding LLM Memory

https://alonsosilvaallende.github.io/blog/posts/2025-06-28-Understanding-Memory/Understanding_LLMs_Memory.html

Using the @marimo_io extension for Quarto by @Posit

New post: Understanding Chat Templates

https://alonsosilvaallende.github.io/blog/posts/2025-06-20-Understanding-Chat-Templates/UnderstandingChatTemplates.html

New post: Understanding Tokenizers
https://alonsosilvaallende.github.io/blog/posts/2025-06-17-Understanding-Tokenizers/UnderstandingTokenizers.html

Celebrating Bell Labs’ 100 years anniversary #NokiaBellLabs100

TIL to do quizzes with JupyterQuiz:
https://alonsosilvaallende.github.io/til/posts/2025-06-16-Jupyter_Quiz/JupyterQuiz.html
I like this simple way to provide feedback to students. I’m using these quizzes to motivate my kid to learn new things.

TIL you can embed a live REPL on a website:
https://alonsosilvaallende.github.io/til/posts/2025-06-15-Embed_REPL/Embed_REPL.html
I think this is a great tool for writing tutorials and blog posts.

My talk "Building Knowledge Graph-Based Agents with Structured Text Generation" at PyData Global is now available on YouTube: https://www.youtube.com/watch?v=94yuQKoDKkE
#PyData #PyDataGlobal

The new release of logits-processor-zoo supports batch inference of the logits processors with vLLM
`pip install logits-processor-zoo`
Here is the repo: https://github.com/NVIDIA/logits-processor-zoo
Here is the commit 😊: https://github.com/NVIDIA/logits-processor-zoo/commit/5d65e369e370f636a3cf6c709b515723e8338aa5

🎵 "Graph RAG enhances RAG-based apps by leveraging structured knowledge graphs with open-weights LLMs." Dive into @alonsosilva 's talk on combining vector and graph retrieval for better context in LLM responses.
🔗 Watch now:https://youtu.be/sjprfQw5TJw

AI Eats the World
https://www.youtube.com/watch?v=LGDa3pO23Wc

Me entrevistaron para el podcast IA para los negocios de EvoAcademy
¡Espero que les guste!
https://youtu.be/-56EfNIGy0k

Mañana presento “Creación de agentes basados en gráficos de conocimiento con generación de texto estructurado y modelos open-weights” en el Centro de Modelamiento Matemático de la Universidad de Chile
https://www.cmm.uchile.cl/?p=59116

Hoy estaré presentando “Mejorando la generación aumentada con recuperación (RAG) con la búsqueda híbrida” en la conferencia EVIC 2024:
https://www.evic.cl/
¡Nos vemos!

Este lunes estaré presentando Potenciando aplicaciones RAG con grafos de conocimiento.
La inscripción es gratuita: https://iadevs.cl/
¡Los espero!

🗣️ "Instead of relying on the Pandas API, we are using Narwhals which provides an abstraction layer over several kinds of tabular data.

This means faster, more efficient handling of tabular data and serious performance gains for data apps at scale."

😍 Love this, from Plotly's blog

💡 Plotly v6 isn't out yet, but you can try it out already with `pip install -U --pre plotly`

https://plotly.com/blog/plotly-dash-major-release/

Stratified cross-validation considered harmful?

Since the early days of @sklearn
we have nudged users into using stratified cross-validation in the presence of class imbalance.

I think this was a design mistake. Here is a notebook to back my claim:

https://gist.github.com/ogrisel/af21bdc55a2c02671a48c68631ee7294

And here is a bluesky summary thread:

https://bsky.app/profile/ogrisel.bsky.social/post/3lcnjg2iv4s2w

🎉 NEW Cookbook: Mistral AI and Solara

- Check out this cookbook on how to create a chatbot with Solara to talk to your PDFs with Mistral AI models: https://github.com/mistralai/cookbook/blob/main/third_party/solara/README.md

Alternatively, you can modify the code directly in PyCafe:
https://py.cafe/alonsosilvaallende/solara-mistral-ai-chat

Big shoutout to @maartenbreddels and Sophia Yang for all the help with this cookbook! 😊

Just gave my talk at PyData Global on Building Knowledge Graph-Based Agents with Structured Text Generation and Open-Weights Models.
Thanks to everyone who attended!
Here is a link to the slides for those interested.
https://alonsosilvaallende.github.io/2024-PyData-Global/