Skip to main content

Experiment Main Metrics History DATA

POST 
/experiments/:experimentId/metrics/main/history

Required Permissions: Experiment Get or Experiment Admin

Request​

Path Parameters

    experimentId int64required

    Possible values: >= 1

    The id of the experiment that you want the history of.

Body

The parameters to filter the history by.

    filters object
    segments object[]
  • Array [
    • segment_id int64

    Possible values: >= 1

    operator string

    Possible values: [in, not_in]

    values string[]
  • ]
    • period string

    Possible values: [1h, 1d, 1w, 1M]

    tz_offset integer

Responses​

A JSON Object with main metrics history per period. The object has three properties, columnNames, columnTypes and rows. columnNames is an array of strings which identifies which columns are which in the rows[][]. columnTypes is an array of strings which identifies the types of data in the rows[][]. rows is an array of arrays which contain the data of the experiment variants over your selected period. The column names are defined as the following:

Regular Columns:​

  • date - The date of the row's data in seconds since epoch.
  • variant - The variant that this row describes. (0, 1, 2 or 3).
  • first_exposure_at - The date of the first exposure of this variant in the current time period in milliseconds since epoch.
  • last_exposure_at - The date of the last exposure of this variant in the current time period in milliseconds since epoch.
  • unit_count - The number of units that were exposed to this variant in the current time period.
  • cum_unit_count - The cumulative number of units that were exposed to this variant at the end of the current time period.

Dynamic Columns:​

  • metric_${metric_id} - The value of the metric with the id of ${metric_id} in the current time period.
  • metric_${metric_id}_mean - The mean value of the metric with the id of ${metric_id} in the current time period.
  • metric_${metric_id}_var - The variance of the metric with the id of ${metric_id} in the current time period.
  • metric_${metric_id}_pvalue - The p-value of the metric with the id of ${metric_id} in the current time period.
  • metric_${metric_id}_impact - The impact of the metric with the id of ${metric_id} in the current time period.
  • metric_${metric_id}_impact_ci_lower - The lower bound of the confidence interval of the impact of the metric with the id of ${metric_id} in the current time period.
  • metric_${metric_id}_impact_ci_upper - The upper bound of the confidence interval of the impact of the metric with the id of ${metric_id} in the current time period.
caution

The columns may come in a different order than the one listed above. It's important to use the columnNames array to identify the index of the column you want to use.

Schema
    columnNames string[]
    columnTypes string[]
    rows array[]
Loading...