I  Files



  • Clear Session (Ctrl+N)


  • "Clear Session" allows the user to start a new session (and clear the old one), meaning that everything that was done before is cleared and NOT SAVED.

  • Open (Ctrl +O)


  • With "Open" the user can load an existing model.

  • Save (Ctrl +S)


  • Save allows the user to save a model. By "model" it is meant ModelIn and ModelOut structures. If logged as an advanced user, the user is asked if he wants to save also the X structure.

  • Import data from .mat file (Ctrl +I or Ctrl +B)




  • The function "import" allows the user to import 2 kinds of data:
    - Single set means that only one matrix for X and one matrix for Y can be loaded.
    - Multiple sets means that the user can load several matrices both for X and Y.
    In the table below, the user can find what kind of data is supported par each method in this version of CuBatch:


     
        Single Set Multiple Set Missing data Data set ref
        1D 2D 3D ND 1D 2D 3D ND    
                           
    Factor Analysis PCA no yes yes yes no yes no no no Matrice.mat
    OPA no yes no no no no no no yes VUBdatatest.mat
    OPA 3D no no no no no yes no no yes VUBdatatest.mat
    Tucker no yes yes yes no no no no yes donpolO3.mat
    Parafac no yes yes yes no no no no yes fluorescence.mat
    Parafac2 no no no no no yes no no yes VUBdatatest.mat
                           
    Regression X PLS no yes yes yes no no no no yes fluorescence.mat
    N-PLS no yes yes yes no no no no yes fluorescence.mat
    IV-PCA no yes no no no no no no no donclim.mat
    IV-Tucker no yes yes yes no no no no no donclim.mat
                           
    Regression Y PLS yes yes1 no no no no no no yes2 fluorescence.mat
    N-PLS yes yes1 no no no no no no yes2 fluorescence.mat
    IV-PCA yes yes no no no no no no no donpolO3.mat
    IV-Tucker yes yes yes yes no no no no no donpolO3.mat

    1 the algorithm remains (n)PLS1: if Y is a matrix, its columns will automatically be treated separately

    2 as the Y is treated column-wise, a sample/batch having a missing value for a certain Y-variable will be skipped in phase of modelling as well as validation whenever a model concerning that variable is being computed

  • Export




    • Data to .mat file (Ctrl +E)


    • Exports all data that were loaded as a *.mat file. These data are automatically saved in the folder "Data" in the common CuBatch directory.
      If both X and Y are present it is asked initially from which one (or both) one desires to export the data.
      A subsequent requester asks whether one wants to export the whole array(s) - "data set" - or only part of it.
      In the figure that opens after it is possible to choose samples/batches and variables in one or both of the arrays in memory. Once the choice is made the user shall just press proceed and select the file to export to. The two variables, when present will automatically be named
      'X' and 'Y' (the "internal" data set name is not affected).

    • Plot (Ctrl +W)


    • Exports plots as jpg or bmp or whatever. These plots are automatically saved in the folder "Plots" in the common CuBatch directory.

    • Report to .txt file (Ctrl +R)


    • Exports informations about data, algorithms, settings, results...etc as a *.txt file. These reports are automatically saved in the folder "Reports" in the common CuBatch directory.

  • Print (Ctrl +P)


  • Prints the current CuBatch window.

  • Exit (Ctrl +Q)


  • Exit CuBatch (and not Matlab)

     

    Back to Introduction Go to Edit