Skip to main content Skip to complementary content

Using Qlik AutoML connections in visualization expressions

You can use the analytic connections in visualization expressions. The syntax when using the analytic connections in expressions must follow the server-side extensions syntax.

Data should be processed in a visualization expression only when it will dynamically change based on the input from a user. If the transaction from the data model will always be the same, the prediction should instead be calculated in the load script and cached in the data model.

If the use case is based on user input, you can use an analytic connection and corresponding server-side extension syntax in your chart expression to create interactive charts that visualize data received from the model endpoints.

  1. When editing a visualization, click Expression to enter the expression editor.

  2. In the expression editor, enter an expression in the expression field. The expression must be constructed using the server side extension syntax.

Information noteDepending on the data size and the called machine learning endpoints, the responsiveness of charts containing analytic connections could be impacted, since data is sent to and returned from Qlik AutoML for calculations.

Working with the expression editor

Server side extensions syntax


Here is an example of a call made to a Qlik AutoML deployment that provides banking customer churn prediction.

sum(aggr(endpoints.ScriptEvalEx('SNNNNNNNNNNSSNNSNNNNNSSSSSNSNNSS','{"RequestType":"endpoint", "endpoint":{"connectionname":"Qlik_AutoML_Churn"}}', id_loan, CurrentBalance, loan_age, delq_sts, Margin, countLatePayment, RefinanceRateRelativity, RealGDP, ChangeUnemploymentRate, CurrentLCV, fico, flag_fthb, cd_msa, mi_pct, cnt_units, occpy_sts, cltv, dti, orig_upb, ltv, int_rt + vInterestRateShift as int_rt, channel, ppmt_pnlty, prod_type, st, prop_type, zipcode, loan_purpose, orig_loan_term, cnt_borr, flag_sc, customerFeedback ),id_loan))