We’ve released a new version of the Datagrok platform (0.94.0). This release focuses on the areas of visualization, usability, rich features for the developers, and traditionally on the platform stability.
Visualization and usability improvements
- Many updates in Scatter plot: a lasso tool, axis sliders, data split by marker shape, custom lines by equations; all summarized in this video
- New Histogram features: splines and bands: video
- New Line Chart features: marker shape selection, whiskers: video
- Using “Relative Values” property in combination with the “Stack” property to analyze the distribution of the stacked values: overview
- Many aspects and issues of the Bar Chart addressed: overview
- Search for category names in filters (useful when there are many categories in the column): more in this video + an overview
- Lists in parameterized database queries: overview + video; also check the video about parameterized queries in general: link
- Database query caching with cron jobs ability: video
Improvements for developers
- More granular event handling (an example with intercepting conditional color coding settings change: link)
- Creating drag-and-drop objects: overview
- Getting cell colors: overview
- Adding custom machine learning models to Datagrok in R, Python, or from external models deployed in clusters: video, package, help
Enhancements in public packages
- PowerPack and Universal Search — commonly used platform enhancements, currently covering Start Page widgets, Power Search (ability to search for anything from the Start Screen), and search templates support: video
- Updates in the Peptides package, learn more in the video
We have also redesigned our approach to interactive Tutorials. We have moved them to a standalone package and made them available through a dedicated app. Check this video for an overview. Every tutorial on the platform now is a simple TypeScript code, like here. Now it is possible to equip customers’ apps and deployments of Datagrok with tutorials as well!
Give the new version a try at https://public.datagrok.ai.