☀️🌖

Welcome 👋

We are a Machine Learning Studio, specialized in Nowcasting/Forecasting, based in Berlin.

We aspire to:

>> Build the most usable, open-core Time-Series Machine Learning tools available. >> Bring speed, rigour and transparency to the field of Forecasting/Nowcasting. >> Work with experts to develop useful tools and solutions to real-world problems. >> Make state of the art Time-Series models accessible to everyone.

Mark Aron Szulyovszky

Daniel Szemerey

Community

PACKAGES 👇

Krisi

Lightweight Evaluation of Time Series Forecasting (both Classification and Regression) with powerful static (pdf) and dynamic (html) Reporting. Supports evaluation of metrics over time.

Fold

A Time Series Cross-Validation library that lets you build, deploy and update composite models easily. An order of magnitude speed-up, combined with flexibility and rigour.

Fold-models

Cross-validation and model composition for non-stationary Time Series. Adds support for third-party libraries, eg.: NeuralForecast, StatsModels or XGBoost.

All Packages ➡

LINKS 👇

TOPICS 👇

Uncertainty Quantification

Point forecasts on their own do not provide the business-critical insights you're looking for.

Forecasting / Nowcasting

We live in the world of Complex Dynamical Systems. Forecasting the future requires a different approach to other Machine Learning fields.

Neuro-symbolic Approach

Combining Deep Learning with interpretable symbolic approaches.

Dream Faster AI

UG (haftungsbeschränkt)

🐻 Berlin

☀️🌖