PyData Trójmiasto #38

Where: Łużycka 8B, Gdynia, Tensor Z building. When: 26th of November at 6PM Registration: please register with your full first and last name. Don't forget to bring your ID when entering the event venue. RSVP available until 6pm day before. In any urgent communication please let us know at [email protected] REGISTRATION LINK: https://www.meetup.com/pydata-trojmiasto/events/311852594/ Parking: Free parking located along the street next to the building. Agenda: 18:00 - 18:05 - Event boarding 18:05 - 18:10 - A few words about PyData 18:10 - 18:55 - From Data to Locations: Predicting the Best Spots for Sinsay Stores by Kazimierz Najmajer 19:00 - 19:45 - When did your LLM go stale? Introducing LLMLagBench for Identifying Knowledge Cutoffs in Large Language Models by Piotr Pęzik, PhD 19:45 - Networking & snacks About "From Data to Locations: Predicting the Best Spots for Sinsay Stores" by Kazimierz Najmajer: Currently, Sinsay - a brand of LPP - is in a period of rapid expansion. Every new Sinsay store starts with a question: Where should we open next? In this talk, I'll share how our team at Silky Coders uses data, maps, and machine learning to answer that question across Poland and other European countries. Discover how our application supports business decision-making. Kazimierz Najmajer is a Senior Data Scientist at Silky Coders, where he focuses on classical machine learning, optimization, and generative AI solutions. In addition to his professional work, he has also taught machine learning at Gdańsk University of Technology (GUT). About "When did your LLM go stale? Introducing LLMLagBench for Identifying Knowledge Cutoffs in Large Language Models" by Piotr Pęzik, PhD: https://huggingface.co/spaces/pelcra/llmlagbench Piotr Pęzik is an associate professor at University of Lodz and a head of NLP research at Voicelab. He has authored numerous publications (scholar.google.com/citations?user=PA4u2rcAAAAJ&hl=en) and developed search and exploration tools for spoken and written corpora as well as natural language processing systems for both general and special domains. More recently, he has been involved in developing large-scale authorship attribution systems for online data. Piotr's research interests have also revolved around the formulaicity vs. novelty of natural languages, an issue which is highly relevant to the evaluation of large generative language models such as GPT-3.

Where: Łużycka 8B, Gdynia, Tensor Z building. When: 26th of November at 6PM Registration: please register with your full first and last name. Don't forget to bring your ID when entering the event venue. RSVP available until 6pm day before. In any urgent communication please let us know at [email protected] REGISTRATION LINK: https://www.meetup.com/pydata-trojmiasto/events/311852594/ Parking: Free parking located along the street next to the building. Agenda: 18:00 - 18:05 - Event boarding 18:05 - 18:10 - A few words about PyData 18:10 - 18:55 - From Data to Locations: Predicting the Best Spots for Sinsay Stores by Kazimierz Najmajer 19:00 - 19:45 - When did your LLM go stale? Introducing LLMLagBench for Identifying Knowledge Cutoffs in Large Language Models by Piotr Pęzik, PhD 19:45 - Networking & snacks About "From Data to Locations: Predicting the Best Spots for Sinsay Stores" by Kazimierz Najmajer: Currently, Sinsay - a brand of LPP - is in a period of rapid expans...