Release Notes Model Exporter Version 4 1

From Eigenvector Research Documentation Wiki
Revision as of 17:36, 22 December 2023 by Donal (talk | contribs) (Created page with "Version 4.1 of Model_Exporter was released in December, 2023. For general product information, see Model_Exporter_User_Guide. ==NEW FEATURES / FIXES== * Added support for...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Version 4.1 of Model_Exporter was released in December, 2023. For general product information, see Model_Exporter_User_Guide.

NEW FEATURES / FIXES

  • Added support for exporting CLSTI models to Matlab, Python, and XML (and JavaInterpreter).
  • Added support for exporting ANNDA and ANNDLDA models to Matlab, Python, and XML (and JavaInterpreter). Please contact Helpdesk before exporting ANNDLDA to XML.
  • Changes to JavaInterpreter: Added methods which can be used once model is loaded and before it makes a prediction:
  1. New methods 'String getPredFieldNames();' and 'String getPredFieldDims();' to get the names and dimensions of predicted quantities. Currently only done for model types PCA, PCR, PLS, and CLS.
  2. New methods 'outputs' and 'axisscale'. These return the 'information/outputs' and 'inputdata/axisscale' elements in the exported model XML file and correspond to the model’s calibration Y-block variable labels and X-block variable’s axisscale respectively, returning String[] and double[] respectively.
  3. New method 'String getYhatNames();' which returns either Y-block variable label (regression) or classnames (classification), or model.classification.classids for CLSTI models. See XML's 'information/yhatnames' element.
  • Fixed bug in exported 'Normalize' preprocessing whenever the window field is used and there are excluded variables.