Indicators¶

The indicator classes compute a single scalar value from the results. This scalar value can be used as an indicator of the outcome fo the test.

NormalizedMeanMetricIndicator¶

class performance.driver.classes.indicator.NormalizedMeanMetricIndicator(config)[source]¶

Calculates the average of the metrics of all runs, normalized by the given value of parameters.

# Note that this needs to be defined in the global `config` section
config:

  indicators:
    - class: indicator.NormalizedMeanMetricIndicator

      # The name of this indicator
      name: someIndicator

      # The metric and summarizer to use in <metric>.<summarizer>
      # format
      metric: someMetric.someSummarizer

      # Normalize each test case measurement to the specified parameter
      normalizeto: normalizeAgainst

      # [Optional] You could also use a python expression. The parameters
      # and the global definitions are available as global variables
      normalizeto: 0.8 * normalizeAgainst

NormalizedMinMetricIndicator¶

class performance.driver.classes.indicator.NormalizedMeanMetricIndicator(config)[source]

Calculates the average of the metrics of all runs, normalized by the given value of parameters.

# Note that this needs to be defined in the global `config` section
config:

  indicators:
    - class: indicator.NormalizedMeanMetricIndicator

      # The name of this indicator
      name: someIndicator

      # The metric and summarizer to use in <metric>.<summarizer>
      # format
      metric: someMetric.someSummarizer

      # Normalize each test case measurement to the specified parameter
      normalizeto: normalizeAgainst

      # [Optional] You could also use a python expression. The parameters
      # and the global definitions are available as global variables
      normalizeto: 0.8 * normalizeAgainst

NormalizedMaxMetricIndicator¶

class performance.driver.classes.indicator.NormalizedMeanMetricIndicator(config)[source]

Calculates the average of the metrics of all runs, normalized by the given value of parameters.

# Note that this needs to be defined in the global `config` section
config:

  indicators:
    - class: indicator.NormalizedMeanMetricIndicator

      # The name of this indicator
      name: someIndicator

      # The metric and summarizer to use in <metric>.<summarizer>
      # format
      metric: someMetric.someSummarizer

      # Normalize each test case measurement to the specified parameter
      normalizeto: normalizeAgainst

      # [Optional] You could also use a python expression. The parameters
      # and the global definitions are available as global variables
      normalizeto: 0.8 * normalizeAgainst