Python 3 wrapper for Weka using JPype.
GPL-3.0 License
Bot releases are hidden (Show)
max_heap_size
parameter of the start
method from the weka.core.jvm
module now has an effectPublished by fracpete 3 months ago
is_additional_measure_producer
variable, additional_measures
property and additional_measure(str)
method to the weka.classifiers.Classifier
class, giving convenient access to classes implementing weka.core.AdditionalMeasureProducer
JavaObject.jclasswrapper
JavaObject.jwrapper
is now deprecated and JavaObject.jobject
should be used insteadJavaArray.new_instance
to JavaArray.new_array
Published by fracpete 5 months ago
setup.py
no longer lists python-javabridge with github URL, as it fails to install via pip otherwisePublished by fracpete 5 months ago
create_instances_from_lists
and create_instances_from_matrices
of the weka.core.dataset
module can handle missing values now (None
in case of lists, nan
in case of matrices), as well as being able to force columns to be nominalload_csv_file
to the module weka.core.converters
to provide a more reliable way of loading CSV files compared to Weka's native CSVLoader
converter (uses Python's csv module and then calls create_instances_from_lists
).simple_range
library as dependency to make it easier to generate integer list from range stringsweka.core.utils
module with following methods: correlation, variance, normalizePublished by fracpete over 1 year ago
Published by fracpete over 1 year ago
Configurable
and JSONObject
into configurable-objects librarysubset_results
, rank_results
to class AttributeSelection
(module: weka.attribute_selection
) to give access to cross-validation output in numeric rather textual form (NB: it has to parse the textual CV output).plot_experiments
method (module weka.plot.experiments
) more customizablePublished by fracpete almost 2 years ago
install_packages
(module: weka.core.packages
) now no longer interprets the installation message returned when installing from URL or local zip file as failure to install; also outputs any installation message now in the console; now has flags for fail_fast
mode (first package that fails stops installation process) and whether to return details
(dict per package rather than just a bool for all packages)install_package
(module: weka.core.packages
) can return either a boolean flag of success or detailed informationPublished by fracpete about 2 years ago
install_packages
and install_missing_packages
of module weka.core.packages
now allow a list of package names instead of tuples (name, version), assuming latest
as versionget_jclass
in module weka.core.classes
can handle primitive classes now as well (eg int
-> java.lang.Integer.TYPE
)get_non_public_field
and call_non_public_method
in module weka.core.classes
allow accessing private/protected fields and calling private/protected methods of Java objects, which avoids having to sub-class classes to get public access to them (NB: only works as long as the security manager allows that)split_commandline
method to module weka.core.classes
, which splits a command-line into classname and option list tupleInstances
class (module: weka.core.dataset
) now supports slicingplot_xmlbif_graph
and xmlbif_to_dot
to module weka.plot.graph
for plotting XML BIF graphs generated by BayesNet using GraphVizplot_graph
to module weka.plot.graph
to plot dot or XML BIF graphsPublished by fracpete over 2 years ago
logging_level
parameter to the start
method of the weka.core.jvm
module, enabling the user to turn off debugging output in an easy way (https://github.com/fracpete/python-weka-wrapper3/issues/40)cv_splits
to class Instances
from module weka.core.dataset
to return a list of train/test tuples as used by cross-validationTester
class (module: weka.experiments
) now has an option to swap columns/rows for comparing datasets rather than classifiersSimpleExperiment
class and derived classes (module: weka.experiments
) now have the additional parameters in the constructor: class_for_ir_statistics, attribute_id, pred_target_columnis_installed
(module: weka.core.packages
) now can check whether a specific version is installedpww-packages
entry point to allow managing of Weka packges from the command-line (actions: list/info/install/uninstall/suggest/is-installed)Published by fracpete over 2 years ago
JavaObject.new_instance
in module weka.core.classes
now automatically installs packages based on suggestions if the JVM was started with the auto_install
flag enabled.test_model_once
of class Evaluation
(module: weka.classifiers
) now has the additional parameter store
, which allows the recording of the predictions (necessary for statistics like AUC)Published by fracpete over 2 years ago
create_instances_from_lists
and create_instances_from_matrices
(module weka.core.dataset
) now allow the specification of column names, for input and output variables.Published by fracpete over 2 years ago
DistanceFunction
class (module weka.core.distances
) (thanks to Martin Trat, https://github.com/fracpete/python-weka-wrapper3/pull/39)JavaArray
class (module: weka.core.classes
) now has __str__
and __repr__
methods that output classname and sizePublished by fracpete over 2 years ago
Published by fracpete almost 3 years ago
python-javabridge
, the new name (fork?) of the javabridge
libraryPackage.__str__
(weka.core.packages
module) method now returns a string rather than printing the name/versionto_numpy(...)
methods to Instance
and Instances
classes (module weka.core.dataset
) to make it easy to obtain a numpy array from the Weka datasetPublished by fracpete almost 3 years ago
weka.core.distances
module for distance functions, with DistanceFunction
base classavg_silhouette_coefficient
method to weka.clusterers
to calculate the average silhouette coefficientPublished by fracpete almost 3 years ago
help_for
to weka.core.classes
module to generate a help screen for an weka.core.OptionHandler
class using just the classname.to_help
method of the weka.core.classes.OptionHandler
class now allows to tweak the generated output a bit better (e.g., what sections to output).plot_classifier_errors
(module weka.plot.classifiers
) now plots the diagonal after adding all the plot data to get the right limitsPublished by fracpete over 3 years ago
Package
class of the weka.core.packages
module now has a version
property to quickly access the version which is stored in the meta-data; the metadata
property now returns a proper Python dictionaryweka.core.packages
module: install_packages
to install more than one package, install_missing_package
and install_missing_packages
to install one or more packages if missing (can automatically stop the JVM and exit the process), uninstall_packages
to remove more than one package in one operationPublished by fracpete over 3 years ago
ASEvaluation
class in the weka.attribute_selection
module now offers the following methods for attribute transformers like PCA: transformed_header
, transformed_data
, convert_instance
weka.core.classes.JavaObject
are now serializable via picklecopy_structure
to the weka.core.dataset.Instances
class to quickly get the header of a datasetheader
to the following classes that returns the training data structure: ASEvaluation
, ASSearch
, Associator
, Classifier
, Clusterer
, TSForecaster
weka.core.serialization
have been moved into weka.core.classes
, with the following methods getting the serialization_
prefix: write
, write_all
, read
, read_all
Published by fracpete over 3 years ago
classes.new_instance
method can take an options list now as wellclasses.get_enum
method to return the instance of a Java enum itemclasses.new_instance
method to create new instance of Java classtypeconv.jstring_list_to_string_list
method to convert a java.util.List
containing strings into a Python listtypeconv.jdouble_to_float
method to convert a java.lang.Double
to a Python floattypeconv
renamed methods: string_array_to_list
to jstring_array_to_list
, string_list_to_array
to string_list_to_jarray
, double_matrix_to_ndarray
to jdouble_matrix_to_ndarray
, enumeration_to_list
to jenumeration_to_list
, double_to_float
to float_to_jfloat
weka.timeseries
module that wraps the timeseriesForecasting
Weka packagePublished by fracpete almost 4 years ago