News from Sympathy for Data 2 | Combine

News from Sympathy for Data 2

After a turbulent spring, it is finally time to write about some of the new features of Sympathy for Data and the updates in the ecosystem we made earlier in 2020. Earlier this year, we released Sympathy for Data 2.0.0 which was the first version with two different editions, an open-source and an enterprise edition. In this blog post, we will follow up on some of the most exciting changes introduced in version 2.0.0 and the recently released version 2.1.0.

An old and a new edition of Sympathy for Data

Combine is proud of the open-source history of Sympathy for Data and to be able to offer the product free of charge for everyone. A tradition we plan to continue in the form of an open-source edition of Sympathy for Data. With the introduction of Sympathy for Data 2.0.0, we decided to introduce a new edition, called Sympathy for Data Enterprise, a commercial product built on top of the open-source edition which extends the functionality with enterprise-tailored features and specialized toolkits. From the start, it includes a machine learning, time series and image analysis toolkit. The number of toolkits and the overall enterprise functionality will be extended with future versions.

We will offer the enterprise edition as a subscription-based license, including support, maintenance and continuous updates with new features and functionality. As usual, we are working in a customer-centric way to bring the best and most requested features for the new toolkits as well as the platform to our enterprise customers. To not forget our valued users of the open-source edition, we will also include selected features and improve functionality in this edition.

With the release of Sympathy of Data 2.0.0, we also changed to semantic versioning and a new release cycle which we will describe later. In general, releases of the enterprise edition will follow the releases of the open-source edition.

What is new in Sympathy for Data 2.0.0 and 2.1.0

Time Series Analysis Toolkit (Enterprise only)

The Time Series Analysis Toolkit provides a set of specialized nodes to work with time-series data and to perform signal processing, including dynamic time warping, time-series aggregation (PAA, SAX, APCA), time series transformation between different time and frequency domains, and more generic nodes for finding local min/max and plateaus.

Machine Learning Toolkit (Enterprise only)

The Machine Learning Toolkit provides nodes to describe, analyze, and model data. It provides nodes for features selection, dimensionality reduction and data pre-processing as well as supervised and unsupervised machine learning algorithms, including K-nearest neighbours, feed-forward neural networks, support vector machines, decision trees and ridge regression. On the advanced clustering methods side, the toolkit contains nodes for density-based spatial clustering (DBSCAN) and agglomerative clustering amongst others.

Image Analysis Toolkit (Enterprise only)

The Image Analysis Toolkit provides a set of specialized nodes for working with image data, including functionality such as property extraction from labelled images, stereo image analysis, generating noise as well as de-noising images, object identification using pre-trained Haar-cascades or simple template-matching.
Furthermore, the toolkit contains utility nodes for arranging images, converting figures to images, or converting between different data types in images and nodes for post-processing data, for example for selecting a relevant subset of lines from a Hough transform or matching key points identified in two images.

Indexed ADAF files

From version 2.1.0, we store a structure index in the ADAF files. It makes repeated queries of the ADAFs structure information much more efficient. For operations that primarily query the structure, for example, listing the available signals and their types, this leads to significant speedups.

Cursors for ADAF and Table viewer

In the latest version, we added cursors to the ADAF and Table viewer. It is a quick way to measure the difference, min, max, and mean values of the selected interval. The data between two cursors can also be exported as a CSV file for further analysis elsewhere.

Signal search for ADAF viewer

Version 2.1.0 adds a new view for searching for signals within an ADAF. It makes it more convenient and faster to locate signals when working with ADAF files containing many different rasters.

New figure wizards

With version 1.6.2, we introduced a new way of configuring new figure nodes with the help of a wizard. In version 2.0.0, we added new wizards for creating histograms, heatmaps, image, and timeline plots. Furthermore, the wizard view is now shown by default when configuring an empty figure node.

Undo in calculator

The powerful Calculator node has finally received an Undo/Redo functionality. It is accessible through buttons in the column toolbar as well as keyboard shortcuts. To make room for the extra buttons, the layout received some optimizations and shows each calculation in the list on a single row.


Sympathy for Data can now be installed without requiring elevated permissions (administrator) in all cases.


With the release of version 2.0.0, we changed to semantic versioning (major.minor.patch) with guaranteed backwards compatibility for minor and patch versions. The general plan is to release one major version and one to two minor versions per year.

If you want to know more about Sympathy for Data, its use cases, functionality and the different editions or you have comments and suggestions for the product, don’t hesitate to contact us at Combine ( or check out your webpage

Kontakta oss