Average Processor

The Average processor groups as described by Group by, then calculates averages and outputs one average for each group.

Parameter	Description
Input Types	Table (number), Table (number, accumulate=True)
Output Types	Table(number)
Group by (group_by)	Grouping processors accept a list of property names to group input items into output items. They produce one output group for an empty list. Most processors generate output from input, often creating one output per input (e.g., DSS input results in DSS output of the same size). Some processors reduce output size by grouping input, performing calculations on each group, and producing one value per group. The output size depends on the grouping scheme, controlled by the Group by configuration parameter. In the case of an empty list, the input is considered to be a single group; thus, the output is of size 1 and either N, DS, or TS. If a list of property names is specified, for example ["system_id", "iface_role"], or a single property is specified, for example ["system_id"], we divide the input into groups such that for each group, every item in the group has the same values for the given list of property names. See the standard deviation processor example for how this works. The output type of a processor depends on a value of the group_by parameter; for an empty list, a processor produces a single value result, such as N, DS, or T, and for grouping by one or more properties it returns a set result, such as NS, DSS, or TS.

Parameter

Description

Input Types

Table (number), Table (number, accumulate=True)

Output Types

Table(number)

Group by (group_by)

Grouping processors accept a list of property names to group input items into output items. They produce one output group for an empty list. Most processors generate output from input, often creating one output per input (e.g., DSS input results in DSS output of the same size). Some processors reduce output size by grouping input, performing calculations on each group, and producing one value per group. The output size depends on the grouping scheme, controlled by the Group by configuration parameter.

In the case of an empty list, the input is considered to be a single group; thus, the output is of size 1 and either N, DS, or TS. If a list of property names is specified, for example ["system_id", "iface_role"], or a single property is specified, for example ["system_id"], we divide the input into groups such that for each group, every item in the group has the same values for the given list of property names. See the standard deviation processor example for how this works.

The output type of a processor depends on a value of the group_by parameter; for an empty list, a processor produces a single value result, such as N, DS, or T, and for grouping by one or more properties it returns a set result, such as NS, DSS, or TS.

Flow diagram showing data processing: Input field 'in' to Average calculation, then Output field 'out'.

In the Add Processor window of the GUI, hover over the "Average with Static Input" tooltip for a visual example of how the Average processor functions.

Data processing pipeline showing input data with system_id and Value columns, processed to calculate average Value per system_id, outputting averages for leaf1 as 12 and spine1 as 32.

Example: Average

See standard deviation example. It's the same except we calculate average instead of standard deviation.