Docs

DataAgent (OnPremise)

Data Source Configuration

Metric Filtering

Metric Filtering

This article introduces multiple methods for filtering metrics in datasets.

Quick start

Click "Filter configuration" to enter the dataset filtering configuration interface, then click "Add" to set the filtering conditions for the metric fields used by the agent.
For metric fields, two filtering methods are provided: "Result filtering" and "Detail filtering". Set the relevant filtering parameters according to actual data requirements, then click "Confirm" to execute the filtering.

Applicable scenarios

Metric filtering methods	Applicable scenarios
Result filtering	Filtering is performed based on the aggregated results of metrics, targeting metric aggregation values. For example: filter data where sum (order quantity) is greater than 2000.
Detail filtering	Filtering is performed based on the original values of metrics, targeting the initial detail data in the dataset. For example: filter data where order quantity is less than 5.

Feature overview

Select the filtering type as "Result filtering" or "Detail filtering".
Select an operator and enter the filter value.
Click "Confirm" to complete the data filtering.

Configuration instructions for relevant parameters are as follows:

Parameters	Configuration instructions
Type	Result filtering: Filtering based on the aggregated results of metrics Detail filtering: Filtering based on the original values of metrics
Operator	Supports =, ≠, >, ≥, <, ≤, (not in) range, (NOT) IN, (not) empty, (not) NaN and other operators warning For range and not-in-range operators, you also need to specify whether the left and right boundaries are open or closed.
Filter value	Enter the filter value for the filtering condition

Last updated: 2026.06.02 14:59:34