DataTalks #38 [Online!]: Intro to Time Series Forecasting

Our 38th DataTalks meetup will take place online, and will provide two time series forecasting-related intros! 📈

Location: Zoom!

Language: English!

𝗔𝗴𝗲𝗻𝗱𝗮:
🍕 17:45 - 18:00 — Online mingling, etc.
🔶 18:00 - 19:00 — Intro to Time Series Forecasting with Deep Learning
🧻 19:00 - 19:10 — A short break
🔷 19:10 - 20:10 — Fast and Scalable Timeseries Modelling with Fugue and Nixtla

***

Talk #1: Intro to Time Series Forecasting with Deep Learning

Speaker: Shay Palachy Affek, Data Science Consultant

Abstract: A short intro talk covering neural network approaches to time series forecasting, including practical aspects and brief overviews of state-of-the-art techniques.

Slides: https://docs.google.com/presentation/d/1UUqC-JgK-9JrUU_4GnU3vDssxoo7gZRxTtyHvoMkx9w/edit?usp=sharing

Bio: Shay is a data science consultant, who led DS teams at several startups, runs the Datahack non-profit and teaches at Tel Aviv University. Get in touch at www.shaypalachy.com .

***

Talks #2: Fast and Scalable Timeseries Modelling with Fugue and Nixtla

Speaker: Kevin Kho, Open Source Community Engineer @ Prefect

Abstract: Timeseries modeling has been one of the weak points of the Python ecosystem compared to R. Statistical modeling libraries such as pmdarima and statsmodels are orders of magnitude slower than R, and state-of-the-art algorithms remain challenging to implement. In this talk, we introduce a set of open-source libraries that allow for fast and scalable time series modeling in Python. Using StatsForecast on different python backends for distributed computing like Dask, Ray, and Spark, we will show the participants how to do forecasting at scale and even how to outperform current benchmarks in the R ecosystem.

Using the Fugue abstraction layer, we’ll learn how to port Python and Pandas code to distributed computation clusters with a few lines of code and leverage the power of Dask, Spark and Ray. This will allow the participants to learn how to train millions of time series models in a few minutes.

Bio: Kevin was a data scientist for four years before working on open-source data tooling. He is a co-author of the Fugue project, an abstraction layer for distributed computing. Most recently, he was an Open Source Community Engineer at Prefect, a workflow orchestration management system.

Note: Both talks will be given in English!