A derivative atomic metric is a metric that is derived from submitted atomic metrics using specific calculation logic. For instance, if one submitted atomic metric represents the payment amount and another represents the number of payers, a derivative atomic metric could be created to determine the average payment amount per person. This topic outlines the process for creating a derivative atomic metric.
Procedure
Navigate to the Dataphin home page and single click the top menu bar Development. You will be directed to the data Development page by default.
Navigate to the New Derivative Atomic Metric dialog box by following the steps below.
Within the New Derivative Atomic Metric dialog box, set the required parameters.
Enter information in the Basic Information section.
Parameter
Description
English Name
Adhere to the following naming convention:
Include letters, numbers, or underscores (_).
Limit to a maximum of 64 characters.
Chinese Name
Adhere to the following naming convention:
Accepts any character type.
Limit to a maximum of 128 characters.
Measurement Unit
Choose the appropriate measurement unit for the derivative atomic metric from options such as Currency Unit, Time Unit, Ratio Unit, Rank, Object Quantifier, and Others.
This includes Currency Unit, Time Unit, Ratio Unit, Rank, Object Quantifier, and Others.
Examples of Currency Unit include cent (CNY), yuan (CNY), and dollar (USD).
Examples of Time Unit include day, month, and year.
Examples of Ratio Unit include decimal, percentile, and thousandth.
Rank is denoted as ranking (rank).
Examples of Object Quantifier include household, transaction, and item.
Weight Unit includes ton and kilogram.
Others is available for additional units.
Data Type
Choose the data type for the derivative atomic metric, which includes options such as string, bigint, double, datetime, timestamp, decimal, Text, Number, Datetime Variable, and Others.
Data Domain
Define the subject area to which the derivative atomic metric pertains.
Description
Provide a concise description of the derivative atomic metric.
Input the computation logic in the Computation Logic section.
Decide whether to perform cumulative calculations across different statistical periods, such as summing the payment amounts over the past seven days based on daily payment amounts.
Define the calculation formula using submitted atomic metrics, for instance, crt_amt/usr_cnt to compute the average order amount per user, where
crt_amtis the order amount andusr_cntis the user count.NoteTo add a field to the computation logic editor, simply click on it in the applicable list.
Once all parameters are set, click Save And Submit to proceed.
What to do next
If the project follows a Dev-Prod pattern, you must publish the derivative atomic metric to the production environment. For more information, see the referenced document.