Node instance associated with the learnable model.
Parameters for the Learner used
ClassTag of the type of data stored in node
classifier need to be defined by the user
classifier need to be defined by the user
Label property for users classifier
syntactic sugar to create simple calls to the function
filter out the label from the features
Run k fold cross validation using the training data.
Run k fold cross validation using the training data. The strategy to split the instances can be set to SplitPolicy.random, SplitPolicy.sequential, SplitPolicy.kth or SplitPolicy.manual if the data splitting policy is not 'Manual', the number of folds must be greater than 1. Otherwise it's value doesn't really matter.
number of folds
strategy to split the instances into k folds.
Loads the model and lexicon for the classifier.
Loads the model and lexicon for the classifier. Looks up in the local file system and the files are not found, looks up in the classpath JARs.
The path of the model file
The path of the lexicon file
Simple logback configuration.
Simple logback configuration.
Hopefully this will be discoverable by just typing loggerConfig.[TAB]
Examples: format: OFF
loggerConfig.Logger("org.apache.spark").setLevel(Level.WARN) loggerConfig.Logger().addAppender( loggerConfig.newPatternLayoutEncoder("%-5level [%thread]: %message%n"), loggerConfig.newConsoleAppender )
format: ON
If you have multiple versions/variations of the same classifier, you can set the following variable, in order to save each variation on different files with different suffixes
Node instance associated with the learnable model.
Parameters for the Learner used
This function prints a summary of the classifier
Test with given data, use internally
Test with given data, use internally
if the collection of data is not given it is derived from the data model based on its type
it is the property that we want to evaluate it if it is null then the prediction of the classifier is the default
it is the property that we want to evaluate the prediction against it, if it is null then the gold label derived from the classifier is used
it is the label that we want to exclude fro evaluation, this is useful for evaluating the multi-class classifiers when we need to measure overall F1 instead of accuracy and we need to exclude the negative class
List of Results
Test with the test data, retrieve internally
Test with the test data, retrieve internally
a Results object
Test with real-valued (continuous) data.
Test with real-valued (continuous) data. Runs Spearman's and Pearson's correlations.
The continuous data to test on
Whether to use caching
A windows of properties
A windows of properties
always negative (or 0)
always positive (or 0)
Represents an instance of a learnable model. Each Learnable instance is associated with a node instance in the data model graph.
Type of the data stored in node