- Видео 11
- Просмотров 3 376
PyData NYC
США
Добавлен 17 июл 2023
PyData is an educational program of NumFOCUS, a 501(c)3 non-profit organization in the United States. PyData provides a forum for the international community of users and developers of data analysis tools to share ideas and learn from each other. The global PyData network promotes discussion of best practices, new approaches, and emerging technologies for data management, processing, analytics, and visualization. PyData communities approach data science using many languages, including (but not limited to) Python, Julia, and R.
Yujian Tang - LLM Based Applications - Building Agentic RAG Workshop
LLM Based Applications - Building Agentic RAG Workshop
Speaker: Yujian Tang, CEO of OSS4Al
Yujian Tang started developing software professionally at the age of 16. In college, he studied computer science, neuroscience, and statistics and published machine learning papers to conferences like lEEE Big Data. After graduation, he worked on the AutoML system at Amazon before moving on to build his own companies including a data aggregation app, an NLP API, and his current company - OSS4Al, an organization aimed at providing all developers access to the resources to understand, use, and contribute to the direction and development of Al.
Speaker: Yujian Tang, CEO of OSS4Al
Yujian Tang started developing software professionally at the age of 16. In college, he studied computer science, neuroscience, and statistics and published machine learning papers to conferences like lEEE Big Data. After graduation, he worked on the AutoML system at Amazon before moving on to build his own companies including a data aggregation app, an NLP API, and his current company - OSS4Al, an organization aimed at providing all developers access to the resources to understand, use, and contribute to the direction and development of Al.
Просмотров: 95
Видео
Zain Hasan - Scaling Vector Search in Production Without Breaking the Bank
Просмотров 302 месяца назад
Scaling Vector Search in Production Without Breaking the Bank: Quantization and Adaptive Retrieval Speaker: Zain Hasan, Senior ML Developer Relations Engineer at Weaviate Everybody loves vector search and enterprises now see its value thanks to the popularity of LLMs and RAG. The problem is that prod-level deployment of vector search requires boatloads of CPU, for search, and GPU, for inference...
Art Anderson - Supercharging Real-time Applications with Vector and Graph
Просмотров 412 месяца назад
Embark on a journey connecting key-value, vector, and graph to build real-world applications showcasing Retrieval Augmented Generation (RAG) using semantic search along with recommendation engines, user profile stores, and more. Unlock the full potential of your real-time applications with multi-model databases and see how Python pulls this all together.
Ilia Zlobin - New Machine Learning Paradigm with DSPy: No Prompt Engineering Required
Просмотров 612 месяца назад
Prompt engineering has some limitations in that you have to tailor it to a particular use case and then adjust and support it as your application continues to evolve. With DSPy, you look at the problem from the traditional machine learning perspective where you operate with datasets orchestration, evaluation metrics and hyper-parameters tuning to improve performance on the task. DSPy gives you ...
Shafik Quoraishee - Multi-Modal Machine Learning and Image Visualization
Просмотров 1173 месяца назад
This talk will take a dive into the core theories and mechanisms behind multi-modal AI models, the powerful AI systems that integrate intelligence from diverse data sources such as text, AI, and vision. While we will reference popular applications like GPT Vision, Gemini, and Sora as illustrative examples, our focus will primarily be on understanding the foundational principles that enable capa...
Jim Dowling - LLMOps with Hopsworks
Просмотров 624 месяца назад
Hopsworks is a data platform tailored for machine learning (ML), featuring a Python-centric Feature Store and MLOps capabilities. This modular platform can be used independently as a Feature Store, for managing, governing, and serving models, or for developing and operating feature and training pipelines. Hopsworks allows ML teams to collaborate by providing a secure, governed environment for d...
Erik Bernhardsson - Introduction to Modal
Просмотров 1075 месяцев назад
Modal's platform empowers data/AI/ML teams to develop faster at lower cost, while scaling production workloads to thousands of CPUs and GPUs. More about Erik: erikbern.com/about
Jorn Mossel - Time Series Forecasting
Просмотров 5966 месяцев назад
Time series problems, such as sales forecasting, energy demand, and peak server load predictions, are widespread in data science, yet often overlooked in introductory machine learning courses. This tutorial offers a brief introduction to time series, highlighting how they differ from other machine learning problems. We then demonstrate both traditional and more advanced time series techniques t...
Thomas J. Fan - Time Series EDA with STUMPY
Просмотров 2 тыс.6 месяцев назад
STUMPY is a robust and scalable Python library for computing a matrix profile, which can create valuable insights about our time series. STUMPY is built with Numba, which parallelizes computing on CPUs and accelerates it with GPUs. In this talk, we learn about the matrix profile and the various methods to compute it. Afterwards, we explore how to use the matrix profile in applications such as p...
Nick Lourie - But, Did You Tune the Hyperparameters Enough? - (March 13, 2024)
Просмотров 1607 месяцев назад
But, Did You Tune the Hyperparameters Enough? Speaker: Nick Lourie Imagine you're working on a neural net. You come up with a great idea! Maybe the model learns better if trained on the shortest examples first. So, you order the examples, feed them to the model, evaluate it, and find... the performance got worse! Is the idea bad? Or, did you just need to tune the learning rate? Optimal design a...
Andrew Hoh - Building LLM Applications in Production (LastMileAI)
Просмотров 1037 месяцев назад
Andrew Hoh is the President and Co-Founder of LastMile AI. Previously, he was a Group PM Manager at Meta AI, driving product for their AI Platform. Previously, he was the Product Manager for the Machine Learning Infrastructure team at Airbnb and a founding team member of Azure Cosmos DB, Microsoft Azure's distributed NoSQL database. He graduated with a BA in Computer Science from Dartmouth Coll...
Can't access Colab Notebook
Please share link for notebook
It`s in the QR @0:55
@@laavispamaya Thanks! Yes, please try the QR code