Connection management

We've improved data source connection management in Mode, giving you the ability to disable automatic schema refresh for a connection. From a connection's page, you can now disable schema refresh, or manually refresh the schema.

To view a connection from your Mode organization, select Organization Settings > Data > Connections and select the data source you'd like to modify.

Group management

We've improved your organization's management of groups. Now, when admins need to add a user to a group, the following options are available:

  • Search for a specific group
  • Search for a user within a group
  • Add or remove users from a group
  • View metadata about a group, including number of spaces and data source connections

To view your Groups, admins can navigate to Settings > People > Groups


Python library version updates

We’ve updated the versions for the following Python libraries:

  • fiona 1.8.8 (currently 1.7.13)
  • geopandas 0.5.1 (currently 0.4.0)
  • hdbscan 0.8.23 (currently 0.8.18)
  • igraph 0.7.1.post7 (currently 0.7.1.post6)
  • keras 2.2.4 (currently 2.2.2)
  • lifetimes 0.11.1 (currently
  • matplotlib 3.1.1 (currently 2.2.3)
  • networkx 2.3 (currently 2.2)
  • nltk 3.4.5 (currently 3.3)
  • numpy 1.17.2 (currently 1.11.3)
  • pandas 0.25.1 (currently 0.23.4)
  • pandas_profiling 2.3.0 (currently 1.4.1)
  • patsy 0.5.1 (currently 0.5.0)
  • pyproj 2.4.0 (currently
  • requests 2.22.0 (currently 2.19.1)
  • scipy 1.3.1 (currently 1.1.0)
  • six 1.12.0 (currently 1.11.0)
  • scikit-image 0.14.3 (currently 0.14.0)
  • scikit-learn 0.21.3 (currently 0.19.1)
  • spacy 2.2.1 (currently 2.0.12)
  • statsmodels 0.10.1 (currently 0.9.2)
  • tensorflow 1.13.2 (currently 1.11.0)
  • urllib3 1.25.6 (currently 1.23)

See the full list of supported Python libraries here

New chart type

We've added a new chart type, called Big Values. Select a Big Value chart to:

  • Visualize KPIs and Trends on top of larger datasets
  • Add additional formatting to customize chart look and feel
  • Apply filters and sort Big Values

For more information on creating Big Values, check out this getting started video.

Answer more questions with Helix

We’ve launched our high performance in-memory data processing system, Helix. Now every query you run in Mode will automatically stream data into our responsive data engine, so that you and your team can visually explore results up to tens of millions of rows.

With Helix, data teams will spend less time on up-front aggregation and calculation — this is no longer a two-step architecture. You can now consolidate multiple queries into one single query to return a larger result set to then slice and dice for additional dimensionality.

Additionally, users can now build reports that rely on filters, rather than parameters for column-specific entries. This reduces database load previously created by parameter use, and empowers a more flexible and fluid report.

Updates to query results sizes

Mode Studio users can view, chart, export, and access the Notebook for query results up to 10 MB.

For Mode Business and Enterprise customers, we offer different plans that support your specific use case. Talk with our customer success team directly to learn more about Helix.

Support for Databricks

We’ve added support for Databricks as a datasource in Mode. To connect to Databricks, first navigate to the Databricks cluster detail page > Advanced Options > JDBC/ODBC tab. Take down the following information:

  • host name
  • port (defaults to 443)
  • database name
  • an authentication token created specifically for your Mode connection (see documentation for more info)
  • the "HTTP path" field

For help finding the settings in the Databricks cluster, check out the Databricks documentation.

In Mode, select Connect a Database in the left hand navigation. In the dropdown list, select Databricks Spark. From there, complete the fields with the above information. Once connected, you can build and share visualizations created on Databricks data.


Please note: users will need to use HiveQL when querying Databricks data in Mode.

If you’d like to connect Mode to a Databricks instance on your internal network, you can use a Bridge Connector.

Report Clone option in Query Editor

We've fixed a bug where previously, the clone option in the Query Editor was not appearing as expected. Now, the clone option appears in the Report menu within the Query Editor view.

HTML Editor Update functionality

We've fixed a bug where previously, the Update button in the HTML Editor would cause the editor, toolbar, and portions of the Left Navigation Panel to disappear from view. Additionally, a loading spinner would incorrectly appear in the lower part of the view.

Now, pressing the Update button no longer incorrectly impacts these components.

Min/Max controls in HTML Editor

We've fixed a bug where previously, the min/max controls in the HTML Editor were not operational. Now, they function in a manner that is identical to that within the Query Editor.

Query history

We've fixed a bug where previously, only failed runs would show in the query history within the Query Editor. Now, both failed and successful runs appear providing a complete record for the user.