Once you have created a connection, you can select data from the available tables to send to Qlik Predict for calculations, and then load that data into your app. This connection can not only be used in your data load script, but also in chart expressions to call model endpoints and perform real time chart expression calculations.
Information noteIn order for a deployment to appear in the ML deployment dropdown, you must select Enable real-time API access in the Create ML deployment dialog window when creating the deployment.
Configurable settings
The following settings can be configured in the connection dialog:
Configurable settings in the connection dialog
Field
Description
Connection
ML deployment: The ML deployment that will be employed to produce predictions.
Information noteThis dropdown list can display a maximum of 100 deployments.
Alias
Specify the model alias referencing the model to use for predictions. Case-sensitive. The default value is the Default alias contained in the ML deployment.
Response Table
Name of Returned Table: Name of the returned table from the deployed machine learning model.
Include SHAP: If checked, the SHAP values will be included in the response.
Include Apply dataset: If checked, the source data will be included in the response.
Include Errors: If checked, the response will include a column with the reason why the prediction was not produced.
Association
Association Field: The name of the field in the source data table containing a unique identifier to use as an index in the response data. The column will be included with its original name and values. This is intended to allow the user to join results with source data. This can be any field with a unique ID, either from the source data or as part of the table load process.
Name
The name of the connection. The default name will be used if you do not enter a name.
Creating a new connection
Do the following:
To create a Qlik Predict connection, machine learning endpoints must be turned on for the Qlik Cloud tenant. For more information, see Enabling ML endpoints in Qlik Cloud.
Under Space, select the space where the connection will be located.
Select Qlik Predict from the list of data connectors.
Fill out the connection dialog fields.
Click Create.
Your connection is now listed under Data connections in Data load editor or Script.
Data connection
In Analytics Services:
Data connections are used to load data from external data sources into Qlik Cloud for the purpose of creating analytics, in the form of apps and scripts. Data connections can load data from databases and remotely stored files.
Activity centers are the central point of access for apps, spaces, and other content. There are four activity centers in Qlik Cloud: Insights, Analytics, Qlik Talend Data Integration, and Administration. Activity centers were formerly known as hubs.
A Qlik Sense or QlikView app: Apps are task-specific, purpose-built applications. Apps contain data loaded from data sources that is interpreted through visualizations.
The Qlik Sense Mobile app: A mobile app for iOS and Android devices. In the mobile app, you connect to and interact with your cloud data. You can work with your available apps.
Data connections are used to load data from external data sources into Qlik Cloud for the purpose of creating analytics, in the form of apps and scripts. Data connections can load data from databases and remotely stored files.
The load script is a sequence of statements that defines what data to load and how to link the different loaded tables. It can be generated with the Data manager, or with the Data load editor, where it also can be viewed and edited.
Charts are objects where calculations, aggregations, and groupings can be made. Graphical visualizations, such as bar charts and pie charts are common examples, but also non-graphical objects such as pivot tables are charts.
A chart consists of dimensions and measures, where the measures are calculated once per dimensional value. If the chart contains multiple dimensions, the measures are calculated once per combination of dimensional values.
A reference pointing to a model within an ML deployment. Predictions from ML deployments are directed to an alias rather than a specific model. This allows flexible and dynamic prediction workflows where models can easily be replaced without requiring configuration or API call updates. An ML deployment can contain up to 10 aliases, including the default alias.
