To receive notifications about scheduled maintenance, please subscribe to the mailing-list You can subscribe to the mailing-list at

The Random Forest Update

  • NEW FEATURE: implemented first machine learning algorithm into Amp: random forest (found in Gap-filling)
  • NEW FEATURE: for the overview plots, it is now possible to restrict the shown data to a previously selected time range, e.g. 1-31 August 2019, with options in the upper right of the VARIABLES tab.
  • NEW FEATURE: Amp now uses .filetype files that store settings for reading data files, this way new filetypes can easily be added by adding a new .filetype file to a specific folder.
  • NEW FEATURE: the project output directory can now be selected via the new top menu. If no output folder was selected prior to exporting data, a default folder will be used instead.
  • CHANGED: output directory is not immediately created by default, but only when something is exported
  • CHANGED: added new menu drop-down menus in top menu bar that contains Load File, Add File and File Settings buttons
  • CHANGED: limit lines, e.g. for showing outlier removal limits, are now again displayed as lines without markers
  • IMPROVED: the window to select filetypes shows info about the filetypes in tooltips
  • IMPROVED: the import of several Events should now work as expected. Each event is imported as new (boolean) column.
  • IMPROVED: it is now possible to add a custom prefix to the output file name
  • IMPROVED: sanitized the code here and there b/c else I go crazy
  • IMPROVED: the SearchBox now also searches in units, not only variable names
  • FIXED: several bugs were removed, and I am sure several others introduced
  • PYTHON ENV: new conda env is AMP2, with updated versions of scikit and the pydot package