What's New in ODYSSEE Explore 2025.1

ODYSSEE Explore completes the ODYSSEE portfolio (ODYSSEE CAE, ODYSSEE A-EYE, ODYSSEE SOLVER) proposing data visualization and data mining charts that allow: 

• To understand if the data is ready to be exploited by the Machine Learning software. 
• To interpret the results with different visions.

Please note ODYSSEE Explore requires an ODYSSEE_Explore feature license for the GUI and an ODYSSEE_Solver feature license to run data exploration graphs with Quasar solver. ODYSSEE Explore does not embed its own Quasar executable in the setup. The user must already have installed ODYSSEE SOLVER, ODYSSEE CAE or ODYSSEE A-EYE.

• Data Mining panels:
  • New Independent Component Analysis (ICA) panel: This computational method separates a mixed signal into statistically independent sub-components. More specifically, a signal is analyzed to see if it is possible to separate these contributing sources from the total observed signal.
  • New Reliability panel: It performs FORM/SORM analysis. FORM (First Order Reliability Method) and SORM (Second Order Reliability Method) are both techniques implemented in Quasar, used in the field of structural reliability analysis to estimate the probability of failure of a system and the reliability index (β) around a targeted point for a Y-constraint value defined as limit of failure.

Figure 1 : Reliability panel

• • Correlation/Covariance plots:
    • Loading time and display have been specifically improved.
    • Zoom functionality is available for both Correlation and Covariance plots.
  • PCA panel, heatmap option:
    • Ability to define the minimum / maximum values of the color scale.
    • Option to select the background color from a wide choice.
    • Option to display only the contour lines instead of the background color.
• Chart panels
  • Visualization menu has been renamed to Charts for better clarity.
  • 2D/3D plot improvements:
    • Possibility to use data from different sources in a single 2D/3D plot.
    • Axis labels for 2D and 3D plots are automatically displayed and populated with column labels from the source file.
• Panels generalities:
  • An icon has been added in the upper left part of the panels to access the settings part. Long-clicking on the top bar of panels to access it still works.
  • Display settings of panels are now accessible with a button at the bottom right of each panel.
• General:
  • General performance improvement when handling big datasets used in several panels.
  • At the startup of an existing project, Explore will check if there is a more recent backup file and suggest using it. It is useful if connection is lost.
  • An explicit error message is now displayed in panels if the quasar path is not correct.
  • Help description has been added for all panels.
• Bug fixes:
  • In the distribution plot, a value from the source file was not included in the distribution histogram plot. This has been fixed.
  • In the Kmeans panel options, it was possible to set a negative value for the number of clusters. This has been fixed.
  • In PCA plots, the last selected point was not displayed. This has been fixed.
  • In the Correlation/Covariance panel, it was not possible to calculate if Data source2 (Optional) was filled in. This has been fixed.