For more than a year, databricks AI/BI democratizes business intelligence and analytics throughout your organization with highly visual and interactive low -code Dashboards. After a short mention in our latest AI/BI blog, we are excited to show the ability to define an expressive, repeatedly usable calculations in Dashboards AI/BI. Calculations allow you to model your data in a more expressive way at the peak of source data sets and create richer, interactive and more business panels.
Calculations are defined by the known SQL syntax so there is no learning curve to use them. It is important that own calculations also allow you to define aggregations and transformation into the upper part of the Dashboard data sets without editing the original queries to the data file. Come in two forms:
- Measures There are aggregated calculations that can be dynamically applied across different groups according to different visualizations. Example:
- Calculated dimensions There are non -agregated calculations that are of order applications. They are useful for formatting or transforming individual line values. Example:
Why use your own calculations?
A reduction in the data file
Suppose you have the following data file:
You want to imagine a profitable range by region. Without your own calculations, you would have to create a new data file with a column of edges:
Although this approach works, the new data file is static and can only support one visualization. The filters applied to the original data file do not affect widgets using new data sets without further manual adjustments.
With your own calculations, you can express a profitable range as aggregation using the formula:
Although this measure is defined on the original monster data set, it is dynamic. When used in visualization, it is automatically updated to reflect the groups and filters applied to the data file. In this example, we can even use the same own calculation for calculating the margin for the area in one visualization and a profitable range for the product in another visualization. Without its own calculations, this would require at least two other “tailor -made” data sets defined by the right groups.
Until now, use has indicated that the authors of the dashboard need Less than half Like many data sets to support the same number of visualizations in the use of habits.
A more united interactive filtering
Interactivity is a key part of what AI/BI instrument panels make. However, interactive filtering via filter widgets and cross filtering works on the basis of a dataset, which means that interactivity intervenes when visualizations are split across Mayra data sets. In such cases, users may need to take further steps to filter all required widgets.
As mentioned above, the consolidation of data sets, as mentioned above, allow the same filters in force across other visualizations. This results in more complicated, intuitive experience – widgets based on the same basic data are easier to configure to respond to the same filters.
Races are also supported in static filters defined in the time. You can read more about interactivity in AI/BI dashboards here.
Expressive
Calculations include support for more than 40 different features and syntax of expressions, basic arithmetic, aggregation, terms of conditioned cases, data casting, and many strap/date handling functions.
These functions allow more than what is already offered by visualization editors. For example, you can:
- Construct any strings using Concat and Concat_WS functions. Example:
- Data format using DATE_FORDAT. Example:
- Own chains of own category using expressions of cases. Example:
Simplified data file authorization
Introducing your own calculations to increase experience with the author of the data file in many ways:
- The actual transformation can be silenced To well -marked your own calculations, keep the mess outside the data file text.
- In the view of the new data file scheme Details of what fields are available for links in calculations and their data types.
- Immediate validation terms and data type inference The calculated customer side is available for its own calculations.
- Semantically consists of a set of data sets It is easier for your dashboard to achieve when the data file is less growing. Having less, more coherent data sets minimizes confusion and other “detective work” to reappear how your data sets are structured every time you need to update data sets.
Presented
Widgets using customs are subject to the same performance optimization as all other Widgets in AI/BI dashboards. This included questions about questions and a quick calculation on the customer’s side for small data sets.
In addition, it has fewer data sets that perform less complex grouping operations, leading to faster queries on data sets, which accelerates the time of iteration when creating and shortening the load time.
How to create and use your own calculations
To create and use your own calculation:
- Click on the reception data file Own calculation.
- In the panel that opens, fill in Name and Expression Field using a data set View the diagram and verify the expression of assembly to guide your expression. Optionally, the enrollment has As Describing what your calculation is doing. Click CREATE.
- Now you can calculate your own calculation in the data file diagram display and select it as well as select any other field in the visualization editor.
Full instructions can be found on the calculations of the customs documentation. If you are a newcomer to Dashboards AI/BI, check out this tutorial and start.
Conclusion
The own calculations bring a new level of Expresify and the flexibility of the AI/BI instrument panel experience and allow dashboard consumers unified more interactive experiences. Check out the documentation where you will find more details on how to manage your customs and a complete list of supported features and syntax expressions.
If you read to explore the latest AI/BI, you can choose one of the following:
- Testing: Get practical experience by logging in for a free trial version.
- Documentation: Immerse yourself in detail with our documentation.
- Website: Visit our website and learn more.
- Demos: Watch our demo videos, take tours of products and get practical tutorials to see these AI/BI in action.
- Training: Start with free product training via Databricks Academy
- Ebook: Download the Meets Ai EBook Intelligence
We can’t wait to see what you build using your own calculations, and we appreciate your feedback when we continue to expand this feature – let us know what support for even more expressive calculations you would like to see!