Resample Time Series

Description

Resample values along chosen regular intervals of time

Application

Raw time series data may contain 'gaps' in time to fill by interpolating values around it. You may also want to upsample or downsample your time series data.

This manipulator is often used as pre-processing step ahead of using models (such as the Anomaly Detection Model) which require data with regular time intervals to function.

How to use

This function requires an input Dataset to be selected. This Dataset must have Time as a column. If you have multiple different tests in the same dataset, they must be identified by an ID column.

• Select a Column identifying multiple tests (eg. and ID column).
• Select the Time column, if not already preselected (if your column is labelled Time it will have been automatically selected).
• Select one or multiple columns in the Columns field that you would like to resample values for.
• Select the Resampling method:
  • Linear Interpolation will use piece-wise linear interpolation to resample your data. This is method is capable of handling larger volumes of data. It will work to resampling (1:1) or upsampling, but not for downsampling.
  • Spline Interpolation will third-degree spline interpolation. It will work for any new sampling rate, but will be slower and not able to process large volumes of data.
  • 'Keep every nth row' should be used for fast downsampling on data which is already uniform.

• The New time step between samples that you input will determine whether you upsample, downsample, or resample your data (if you match the original sampling rate).
• Finally, you can choose whether to overwrite the existing dataset or Save the dataset under different name.

Example

In the example below, you can see the top plot containing gaps in the data (irregular intervals of time). The Resample Time Series manipulator transforms this data for each value to be plotted against a regular user-defined time interval.