weka_gui_explorer_Explorer.props

toc

= File =

= Description = This props file determines what schemes and options are initially set in the Explorer.

= Version =
 * > 3.5.3

= Fields = >> if set to true the Capabilities filters in the GOE will be initialized based on the full dataset that has been loaded into the Explorer otherwise only the header >> Lists all the tabs that should be displayed in the Explorer. Apart from the Preprocess panel itself, all other panels are basically plugins. >> See the Adding tabs in the Explorer article for more details on adding custom panels. >> Defines the initial directory for opening datasets in the Preprocess panel. >> The following placeholders are recognized (work across platforms): >>> - the temp directory >>> - the user's home directory >>> - the current directory **(the default setting)** >>> - gets replaced by a single percentage sign >> Enable/disable the creation of undo files (default is enabled) >> Specify the directory to use for saving undo files >> The following placeholders are recognized (work across platforms): >>> - the temp directory >> the filter to use, none if left empty >> the classifier to use >> the default test mode in the classify tab >> the default number of folds for CV >> whether the evaluation of the classifier is done cost-sensitively >> a cost matrix still has to be provided! >> whether the confusion matrix is output >> whether the entropy based evaluation measures of the classifier model are output >> whether the classifier model is output >> whether additional per-class stats of the classifier model are output >> whether the predictions of the classifier output as well >> the default percentage split in % >> whether the order is preserved in case of percentage split >> the default random seed >> whether the predictions of the classifier are stored for visulization purposes >> whether to output Java source code for classifiers that implement the interface >> the default classname of the generated Java source code >> the default classname for the class generating the plot instances of the classifier errors >> the minimum size for the crosses that display the classifier errors for numeric class attributes >> the maximum size for the crosses that display the classifier errors for numeric class attributes >> the clusterer to use >> the default test mode >> whether the clusters are stored for visualization purposes >> the default associator >> the default attribute evaluator >> the default attribute selection search scheme >> the default test mode >> the default number of folds for CV >> the default random seed
 * **Preprocess panel**
 * (> 3.6.0, developer version and snapshots later than 07/02/2009)
 * (> 3.6.5, > 3.7.4 and snapshots later than 09/01/2011)
 * (> 3.6.5, > 3.7.4 and snapshots later than 09/01/2011)
 * **Classify panel**
 * 1 - cross-validation (default)
 * 2 - percentage split
 * 3 - use training set
 * 4 - supplied test set
 * (> 3.5.5)
 * (> 3.5.5)
 * (> 3.7.0)
 * (> 3.7.0)
 * (> 3.7.0)
 * **Cluster panel**
 * 2 - percentage split
 * 3 - use training set (default)
 * 4 - supplied test set
 * 5 - classes to clusters evaluation
 * **Associations panel**
 * **Attribute selection panel**
 * 0 - use full training set (default)
 * 1 - cross-validation

= See also =
 * Properties File