What's new in version 18.5.0?

IBM® SPSS® Modeler adds the following features and enhancements in this release.

  • Many new date_add and time_add database functions are now available in the Expression Builder, allowing you even more flexibility when creating expressions. For more information, see Date and Time Functions.
    Figure 1. Graph Output tab
    Graph Output tab for the Extension Output node
  • With the new Python tab in the Extension nodes, you can now use native Python for scripting. Invoke native Python APIs from your scripts to interact with IBM SPSS Modeler. For more information, see Native Python APIs.
  • The Extension Output node for Python Spark now includes a Graph Output tab like R does. For example, you can now use a script such as the following to generate graphs:
    import spss.pyspark.runtime

import numpy
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

ascontext = spss.pyspark.runtime.getContext()
indf = ascontext.getSparkInputData()
sns.pairplot(indf.toPandas(), hue='K')
sns.pairplot(indf.toPandas(), hue='Age')
plt.show()
    Figure 2. Graph Output tab
    Graph Output tab for the Extension Output node
  • A new setting called Calculate ensemble distribution graph is available in the Auto Classifier and Auto Numeric nodes. The setting is turned off by default, and controls whether ensemble distribution graphs are included in the generated auto model output. Also note that the Calculate predictor importance setting has been changed to not be selected by default. When both settings are turned off, auto modeling performance is greatly improved.
  • RedHat 9 is now supported.
  • macOS Sonora 14.0 is now supported.
  • SUSE 15 is now supported.
  • Oracle 21C is now supported.
  • SQL Server 2022 is now supported.
  • DB2 11.5.9 is now supported.
  • R 4.2.2 is now supported.
  • Apache Spark 3.4 is now supported.
  • Apache Hive 3.1.3 is now supported.
  • Apache Impala 4.0.0 is now supported.
  • Python 3.10.7 is now supported.
  • Apple M1 Pro is now supported.
  • You can now browse the files on OneDrive over network drive from SPSS Modeler Client.
  • File browsing using the WEBDAV is now supported.