Options for Exporting Data

There are multiple ways to get your data out of Riffyn and onto your computer. From small to large:

  • copy to clipboard the highlighted value from any cell throughout the app
  • copy to clipboard the values of multiple highlighted cells in multi-valued data editor (top panel)
  • copy to clipboard the values for selected runs from any one column in the bottom panel (here)
  • export to CSV all single-valued data associated with a run group on a step (here)
  • export to CSV all data for all steps of currently viewed experiment (here)
  • export to CSV all data for all steps for all experiments of currently viewed process (here)

Experiment or process data can also be exported using our API, e.g., via the Riffyn Tools JPM Add-in. The latter includes an interface to export a subset of all experiments of a process.

 

When exporting data for one or more experiments, Riffyn will calculate a single data table that is ready for statistical analysis in a variety of 3rd party applications. Riffyn performs a variety of restructuring calculations to transform experiment data into these tables, including the following:

  • export-only formulas saved on properties are performed for every row of every run
  • multi-valued data is reduced or cleaned if any data cleaning procedures have been specified
  • single-valued data is automatically joined to multi-valued data written to the same run
  • data written to connected steps are joined automatically, either by run connections alone or by custom join rules that use matching data on paired properties to constrain how rows of multi-valued run data are joined across connected runs

These restructuring calculations can be initiated by clicking the data status indicator at any time and as many times as desired, so that Riffyn can generate a downloadable file more quickly at a later time when it is urgently needed.

If Riffyn seems always to be exporting a much larger data table than expected or desired, consider updating the default export method used for all experiments of a process with data cleaning procedures or custom join rules. Use custom join rules to constrain how data should be joined across connected runs on different steps. Use data cleaning procedures to specify how data should be joined when two or more sets of multi-valued data have been written to the same run on the same step.

Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request