Olivier Jeunen

When invited to the Super Data Science Podcast, Ritchie Vink, author of Polars, mentioned an upcoming Polars book set to launch in January 2025: "Python Polars: The Definitive Guide" by Jeroen Janssens and Thijs Nieuwdorp. The official Github repository already has a lot of code examples that worth checking out: https://lnkd.in/dJGbwQnM The Early Release version can also be read on O'Reilly Learning platform: https://lnkd.in/daTEf8Tv The book is offering guidance for you to:  - Process larger-than-memory datasets at record speed - Apply the eager, lazy, and streaming APIs of Polars and decide when to use which - Transition smoothly from Pandas or Spark to Polars - Integrate Polars into your existing codebase - Work with Arrow and Parquet to efficiently read and write data - Translate complex ETL tasks into efficient and elegant queries I started reading the Early Release version and I am enjoying it so far. I am looking forward to the final version of the book. #Python #DataFrameLibrary #Polars

40

2 Comments

in Pandas 2.2 one can set pyarrow as the default backend. in this presentation the advantages of arrow vs numpy are presented: 1. Faster string operations 2. Consistent missing value handling 3. Efficient data serialization Minute 24 https://lnkd.in/gbbGpR56

22

Closing thoughts on PyData Eindhoven, data scientists have a very good understanding regarding good uses cases for Generative AI, and, as well, use cases that don't really fit Generative AI and are better served by other approaches. For those use cases, the discussion is about monitoring, quality control and management of AI applications. My talk was the first talk in the back-up list, however, none of the titular roster speakers needed to excuse him/her self, so it would be next year. For those wondering what my talk was to be about, check the following repo :) https://lnkd.in/ehwC7h4H

12

"Are we living in an AI bubble?"   As someone working in the data industry with a clear focus on AI, this question crosses my mind from time to time. Yesterday, a simple interaction at the supermarket provided a clear answer.   An elderly lady, unable to speak Dutch or English, effortlessly communicated with a cashier using her smartphone. She spoke in her native language, showed the screen to the employee, who then responded in Dutch. The phone translated both ways in real-time, facilitating a smooth conversation. Both seemed happy with the interaction they had just had, which wouldn't be possible without AI technology.   This everyday scene showcases how AI is quietly revolutionizing communication, even for those we might not consider early adopters. It's a powerful reminder that AI's impact extends far beyond tech companies and academia.   Have you noticed similar AI-powered interactions in unexpected places? Share your experiences below!

19

1 Comment

Since I had some spare time during the weekends I put something together: https://lnkd.in/eyjqrzbn This is the start of a new journey where I work on my passion of applying modern ML/NLP/AI/GNN techniques for Source Code (and Binary) Vunerability detection and Reverse Engineering. It features a new site, that I personally built, and where I planning to publish my personal Research and Tools. #NLP, #BERT, #CodeBERT, #GraphCodeBERT, #Transformers, #Cybersecurity

13

Are open LLM models truly open? 🤔 It seems they might not be. Researchers at Radboud University assessed over 40 large language models and six text-to-image generators, including OpenAI’s #ChatGPT and #DALL·E2. They evaluated 14 characteristics concerning Availability (source code, pretraining data, base weights, fine-tuning data, etc.), Documentation (code, architecture, preprint paper, etc.), and Access (downloadable package and open API). Results: Of the language models, OLMo7B Instruct and LLM360/AmberChat were the only fully available. BigScience’s BLOOMZ was the only model that was fully open in documentation. Some prominent “open” models scored less well. Alibaba Group’s Qwen 1.5, Cohere’s Command R+, and Google’s Gemma-7B Instruct were judged closed or partially open for most characteristics. Neither Meta’s Llama 2 Chat nor Llama 3 Instruct achieved any open marks. Larger corporations often claim openness while providing limited transparency compared to smaller entities that adhere more closely to open-source principles. The study highlights the risks of relying solely on licensing to determine openness, as it can be manipulated to meet minimal standards without genuine transparency. It's time regulatory bodies adopt composite and gradient approaches to assess openness. Companies/Businesses adopting LLM models need to scrutinize and check for detailed documentation to ensure that AI systems labelled as open-source genuinely adhere to open-source principles. How can we ensure the AI systems we rely on are truly transparent and accountable? Share your thoughts in the comments. 👇

10

1 Comment

My colleague Jeroen Overschie made me finally understand how you can feed relevant documents into a large language model to enhance its context and provide better information and how important a good, relevant search engine still is for this method. In short: if your search engine and the results it turns up are better, the quality of answers from the LLM will be better.

9

I've talked about this before: Frontier Models > Older Model Methodology/ALgos with (years of data/years of research). See Erik Meijer from Microsoft talking about how frontier LLM doing much better than older research and algos. Other examples: latest and the greatest traditional OCR (Object Character Recognition) vs LLM-based visual recognition in OpenAI's 4o-mini. I've also anecdotally heard of Spotify's existing recommendations ML algo with access to years of music-listening user data giving lower quality recommendations than uploading a screenshot of favourites in Spotify app to an LLM and asking for recommendations. Insightful talk from Erik Meijer (MSFT VB, C#, LINQ): link in comments. The best part title of his talk: "Hinton's Nightmare is My Dream" 🤣

10

5 Comments

We built a Local Chatbot for Belgian Stock Market Data 📊💡 Over the past few weeks, we worked on a project that leverages LLMs and the top 20 Belgian stocks on the market. Collaborating with Sofyan Ajridi and Jiri De Jonghe, we developed a local AI agent that can answer questions about these 20 stocks using daily live stock market data. The idea was to use the power of function-calling tools to build a sort of chatbot that can help you get real/live stock market insights in order to make better trading decisions. All of that done locally (M2). #AI #LLM #Finance - code available on GitHub: https://lnkd.in/e-VBWtC3 - Medium article: https://lnkd.in/eKge4fKJ

143

5 Comments

In PyData Eindhoven definitely the building of trusted Generative AI applications is the main theme. Not only the building of Retrieval Augmented Generation pipelines is being discussed, but also, how to make them performant and traceable. A lot of the discussions made me remember some talks by Dr. Chris Hillman (Who you should definitely follow), and this article by our colleague Chetan Hirapara (Who you should also definitely follow). https://lnkd.in/g5JYrhHX

32

2 Comments

🚀 Discover the Future of Generative AI: Gemini Advance vs. Chat GPT-4.0! 🚀 As a Data Scientist, I'm always exploring the latest advancements in AI to stay ahead of the curve. Recently, I delved into the capabilities of Gemini Advance and compared it with Chat GPT-4.0 to see which one stands out in our rapidly evolving field. 🔍 What's Inside: A comprehensive comparison of Gemini Advance and Chat GPT-4.0 Insights into their unique features and performance My personal take on which model excels and why If you're passionate about Generative AI and eager to know which model I prefer, check out my latest article on Medium: https://lnkd.in/gyiwKG79 I would love to hear your thoughts and experiences with these models! Feel free to share your insights and let's spark a conversation about the future of AI. #GenerativeAI #MachineLearning #ArtificialIntelligence #TechTrends #AIResearch

3

Agents have promised us a lot — from automated software development to handling your communications. Nathan Warren and I wrote about where AI agents are now, and where this is all going.

11

Is Chain-of-Thought (CoT) Always Worth It? Just read an insightful paper analyzing CoT prompting in large language models (LLMs). The takeaway? CoT mainly boosts performance on math and logic tasks, with little benefit for other types of problems. The study found that on benchmarks like MMLU, directly generating answers without CoT produces nearly the same accuracy—unless the task involves symbolic reasoning (like equations). CoT’s gains come from handling intermediate steps, but symbolic solvers often perform better. Is this paper proves what we are seeing with o1 models ? The extra cost and time during inference is not worth it in most situations but will help when it comes to hard reasoning prompts https://lnkd.in/dm7PY45n #AI #MachineLearning #LLM #ChainOfThought #NLP #AIResearch

14

🚀 Proud Moment: Introducing Model2Vec by Thomas van Dongen and Stéphan Tulkens! 🚀 I'm incredibly proud to share that my son Thomas, and his friend Stephan have launched *Model2Vec*, now live on GitHub! 💡What is Model2Vec? Model2Vec is a simple method to distill any complex Sentence Transformer model into a small, fast, high performance static model. It creates embeddings that can be used directly for unsupervised tasks, as well as trained to further improve performance. 💪🏻 Why Support Model2Vec? - Open Source: Encourages collaboration and innovation. - Versatile Applications: Suitable for a wide range of industries. - Community Driven: Your feedback can help shape its future. 🔗 Please check it out and give it a ⭐️ on GitHub: [Model2Vec on GitHub](https://lnkd.in/ePc3pM8j) #Model2Vec #OpenSource #DataScience #AI #ProudParent

27

Looking how to select better representatives from your clusters? Take a look at Lexin Chen's PRIME repo: https://lnkd.in/enM_Btbx Now with updated functionality and tutorials! Great collaboration with Arup Mondal and Alberto Perez! #moleculardynamics #clustering #machinelearning

65

Here's my 2024 LinkedIn Rewind, by Coauthor.studio: GenAI isn't just about models - it's about building platforms that empower teams. 2024 taught me that firsthand as we transformed proofs-of-concept into production at Rabobank. From serving as GenAI Tech Coach to launching our GenAI Store platform, this year proved that real innovation happens when we combine technical excellence with human collaboration: 🏗️ Technical Impact: • Led successful GenAI Store PoC development and launch • Implemented enterprise-grade RAG architecture on Azure • Established robust security and data governance frameworks • Built comprehensive platform monitoring solutions 🤝 Community Growth: • Coached 52 teams (250+ participants) at PLAIGROUND Hackathon • Presented GenAI innovations at Rabo Engineers Week • Contributed to MLOps and GenAI knowledge sharing • Promoted to Senior Machine Learning Engineer Three posts that captured our journey: "Rabo Engineers Week: GenAI Store Launch" On building enterprise-ready AI platforms with security and governance at core "This is just a solid beginning, we're going in full throttle 😎" https://lnkd.in/eHngy_aZ "PLAIGROUND Hackathon: Turning Ideas into Reality" Why innovation requires giving teams space to experiment with AI "52 teams, 250 participants - It's a wrap!!" https://lnkd.in/e5MVpdj4 "Celebrating 1 Year at Rabobank" On continuous learning and community impact "I'm so happy to be working for a bank where I am continuously learning, helping colleagues, challenging and developing myself every single day" https://lnkd.in/evvhEJQj 2025 will be about scaling our GenAI platform and empowering more teams to build responsible AI solutions. The foundation is set - now it's time to accelerate 🚀 Huge thanks to my Gen!Lab team and leaders like Stijn de Koning for the trust and opportunity to drive innovation forward. #MLOps #GenAI #RaboEngineersWeek #LinkedInRewind #2024wrapped -- Get your 2024 LinkedIn Rewind! Go to coauthor.studio

22

The Maximalists are reporting on their experience fine-tuning small language models (SLMs) such as Llama 3(.1) and Qwen-2 with 7/8B and 1.5B parameters calling the result BRAG - reporting good values when high quality data is used https://lnkd.in/ea5sThGk. #smalllanguagemodel #slm #generativeai #artificialintelligence #llama3 #qwen2 #themaximalists

1

Interesting read from NVIDIA: "𝐑𝐚𝐧𝐤𝐑𝐀𝐆: 𝐔𝐧𝐢𝐟𝐲𝐢𝐧𝐠 𝐂𝐨𝐧𝐭𝐞𝐱𝐭 𝐑𝐚𝐧𝐤𝐢𝐧𝐠 𝐰𝐢𝐭𝐡 𝐑𝐞𝐭𝐫𝐢𝐞𝐯𝐚𝐥-𝐀𝐮𝐠𝐦𝐞𝐧𝐭𝐞𝐝" RankRAG is a framework designed to instruct and finetune LLMs for context ranking and answer generation in RAG applications. Link: https://lnkd.in/gsEJ5cVK #llm #nvidia #rag #finetune

7