Multihorizon Heat Predictions & Apache Spark vs. cloud native SQL

# PyData Karlsruhe

Achieving High Business Demands with State-of-the-Art Multihorizon Heat Predictions & Apache Spark vs. cloud native SQL engines

This event will be in English.

## Agenda

18:00 Doors open
18:30 Welcome
18:45 Achieving High Business Demands with State-of-the-Art Multihorizon Heat Predictions - Martin Danner, scieneers GmbH
19:15 Networking with dinner and beverages
20:00 Apache Spark vs. cloud native SQL engines - Franz Wöllert, Heidelberger Druckmaschinen AG
20:30 Lightning Talks
20:45 Networking with snacks and beverages
21:30 End

## Lightning Talks

Join us by contributing a five minute lightning talk! Contact pydatasw@uwekorn.com

## Talks

Achieving High Business Demands with State-of-the-Art Multihorizon Heat Predictions: A Journey from ARIMA Models to Temporal Fusion Transformers at Iqony
Martin Danner, scieneers GmbH

Accurately predicting heat demand is critical to optimizing plant operation in heat generation. While recurrent neural networks are superior to simple ARIMA models, they may not meet evolving business needs, particularly for probabilistic forecasting, interpretability, and capturing cross-site complexity. Data fusion approaches combined with transformer-like architectures can address these challenges by merging data of multiple modalities. In this talk I will illustrate the journey from classic ARIMA models to LSTMs and Temporal Fusion Transformers in the context of precise, multi-horizon heat demand forecasts at Iqony and discuss hurdles to overcome during development.

***

Apache Spark vs. cloud native SQL
Franz Wöllert, Heidelberger Druckmaschinen AG

Currently, SQL and Cloud Data Warehouses (DWH) are extremely popular for good reason. They are great for dashboarding and business intelligence (BI) use cases due to their ease-of-use. However, their combination might not be the best choice for every problem. More precisely, business-critical data pipelines with high complexity might be better suited for frameworks like Apache Spark.

Expect an opinionated comparison between Apache Spark and seemingly easier-to-use cloud native SQL engines. By the end of this talk, you will be challenged to think about why they are complementary and when each has its justification.

***

Lightning Talks:

your spot, ping us!
tba.
tba.

***

## Acknowledgement

Also a big thank you to our sponsors:

• NUMFOCUS, for promoting open source software.
• QuantCo Deutschland GmbH, for supporting the organization.
• Königsweg as Founding Sponsor
• scieneers GmbH, for hosting the meetup.

## Contact

If you have any questions or suggestions, please feel free to contact us via:

• Meetup
• pydatasw@uwekorn.com