Public/Start-NexosisForecastSession.ps1
|
Function Start-NexosisForecastSession { <# .Synopsis Start a forecast session for a submitted data source using the Target Column, Columns Meta-data, and a range to forecast. .Description Forecast sessions are used to predict future values for a data source. To create a forecast session, specify the data source to forecast, as well as the start and end dates of the forecast period. The Nexosis API will execute a series of machine learning algorithms to approximate future values for the data source. The forecast start date should be on the same day as (or before) the last date in the data source. If there is a gap between your forecast start date and the date of the last record in your data set, the Nexosis API will behave as if there is no gap. .Parameter dataSourceName Name of the data source (view, dataset, etc) to forecast .Parameter targetColumn Column in the specified data source to forecast .Parameter startDate First date to forecast date-time formatted as date-time in ISO8601. .Parameter endDate Last date to forecast date-time formatted as date-time in ISO8601. .Parameter resultInterval The interval at which predictions should be generated. Possible values are Hour, Day, Week, Month, and Year. Defaults to Day. .Parameter callbackUrl The Webhook url that will receive updates when the Session status changes If you provide a callback url, your response will contain a header named Nexosis-Webhook-Token. You will receive this same header in the request message to your Webhook, which you can use to validate that the message came from Nexosis. .Example # Start a Daily Forecast session on the data source 'salesdata' and set the target column to forecast to 'sales' - forecast between the range of 01-06-2013 to 01-13-2013 Start-NexosisForecastSession -dataSourceName 'salesdata' -targetColumn 'sales' -startDate 2013-01-06 -endDate 2013-01-13 -resultInterval Day -columnMetadata $columns #>[CmdletBinding(SupportsShouldProcess=$true)] Param( [Parameter(Mandatory=$true, ValueFromPipeline=$True)] [string]$dataSourceName, [Parameter(Mandatory=$false, ValueFromPipelineByPropertyName=$true)] [string]$targetColumn, [Parameter(Mandatory=$true, ValueFromPipelineByPropertyName=$true)] [DateTime]$startDate, [Parameter(Mandatory=$true, ValueFromPipelineByPropertyName=$true)] [DateTime]$endDate, [Parameter(Mandatory=$false, ValueFromPipelineByPropertyName=$true)] [ResultInterval]$resultInterval=[ResultInterval]::Day, [Parameter(Mandatory=$false, ValueFromPipelineByPropertyName=$true)] [string]$callbackUrl, [Parameter(Mandatory=$false, ValueFromPipelineByPropertyName=$true)] $columnMetadata=@{} ) process { if (($dataSourceName -eq $null ) -or ($dataSourceName.Trim().Length -eq 0)) { throw "Argument '-DataSourceName' cannot be null or empty." } if ($columnMetadata -isnot [Hashtable]) { throw "Parameter '-columnMetaData' must be a hashtable of column metadata for the data." } $params = [System.Web.HttpUtility]::ParseQueryString([String]::Empty) $params['dataSourceName'] = $dataSourceName if (($null -ne $targetColumn) -and ($targetColumn.Trim().Length -ne 0)) { $params['targetColumn'] = $targetColumn } if ($startDate -ne $null) { $params['startDate'] = $startDate } if ($endDate -ne $null) { $params['endDate'] = $endDate } if ($callbackUrl -ne $null){ $params['callbackUrl'] = $callbackUrl } $params['resultInterval'] = $resultInterval.toString() if ($pscmdlet.ShouldProcess($dataSourceName)) { if ($pscmdlet.ShouldProcess($dataSourceName)) { $response = Invoke-Http -method Post -path "sessions/forecast" -Body ($columnMetadata | ConvertTo-Json -depth 6) -params $params -needHeaders $responseObj = $response.Content | ConvertFrom-Json $responseObj } } } } |