Definitions for a Weka J48 and PART classifiers. More...

#include <config.h>
#include <stdlib.h>
#include <stdio.h>
#include <assert.h>
#include <inttypes.h>
#include <unistd.h>
#include <string.h>
#include <errno.h>
#include <libxml/tree.h>
#include <libxml/parser.h>
#include <libxml/valid.h>
#include <jwsc/base/error.h>
#include <jwsc/base/file_io.h>
#include <jwsc/vector/vector.h>
#include <jwsc/matrix/matrix.h>
#include "xml.h"
#include "haplo_groups.h"
#include "weka.h"

Functions
static Error *	read_weka_model_labels (char *labels_str_out, const char labels_fname)
	Reads a list of Weka model labels.
Error *	train_weka_j48_model (const Vector_u32 labels, const Matrix_i32 markers, const char labels_fname, const char model_fname, const char *weka_jar_fname)
	Trains a Weka J48 model and writes it to a file.
Error *	train_weka_part_model (const Vector_u32 labels, const Matrix_i32 markers, const char labels_fname, const char model_fname, const char *weka_jar_fname)
	Trains a Weka PART model and writes it to a file.
Error *	predict_labels_with_weka_j48_model (Vector_u32 labels_out, Vector_d confs_out, const Matrix_i32 markers, const char labels_fname, const char model_fname, const char weka_jar_fname)
	Predicts the labels for a set of marker samples using Weka's J48 classifier.
Error *	predict_labels_with_weka_part_model (Vector_u32 labels_out, Vector_d confs_out, const Matrix_i32 markers, const char labels_fname, const char model_fname, const char weka_jar_fname)
	Predicts the labels for a set of marker samples using Weka's PART classifier.
static Error *	read_weka_xml_doc (xmlDoc *xml_doc_out, const char xml_fname, const char *dtd_fname)
	Reads and and optionally validates an XML document.
static Error *	create_weka_model_tree_from_xml_node (Weka_model_tree *tree_out, const Weka_model_tree parent, uint32_t parent_label, xmlNode xml_node, const char model_dirname)
	Allocates and initializes a model tree from an xml node.
static Error *	create_model_training_data (Vector_u32 train_labels_out, Matrix_i32 train_markers_out, const Vector_u32 data_labels, const Matrix_i32 data_markers, const Vector_u32 model_labels, Vector_u32 const *model_altlabels)
	Creates a set of training data as a model-specific labeled subset of a larger data set.
static Error *	train_weka_j48_model_node (Weka_model_node node, const Vector_u32 labels, const Matrix_i32 markers, const char weka_jar_fname)
	Trains a model in the tree using node-specific data.
static Error *	train_weka_part_model_node (Weka_model_node node, const Vector_u32 labels, const Matrix_i32 markers, const char weka_jar_fname)
	Trains a model in the tree using node-specific data.
Error *	train_weka_j48_model_tree (Weka_model_tree *tree_out, const Vector_u32 labels, const Matrix_i32 markers, const char tree_xml_fname, const char tree_dtd_fname, const char model_dirname, const char *weka_jar_fname)
	Trains a Weka J48 model tree .
Error *	train_weka_part_model_tree (Weka_model_tree *tree_out, const Vector_u32 labels, const Matrix_i32 markers, const char tree_xml_fname, const char tree_dtd_fname, const char model_dirname, const char *weka_jar_fname)
	Trains a Weka PART model tree .
static Error *	recursively_predict_j48_labels_in_model_tree (Vector_u32 labels_out, Vector_d confs_in_out, const Weka_model_tree tree, const Matrix_i32 markers, const char *weka_jar_fname)
	Recursively predicts a label from a model tree.
static Error *	recursively_predict_part_labels_in_model_tree (Vector_u32 labels_out, Vector_d confs_in_out, const Weka_model_tree tree, const Matrix_i32 markers, const char *weka_jar_fname)
	Recursively predicts a label from a model tree.
Error *	predict_labels_with_weka_j48_model_tree (Vector_u32 labels_out, Vector_d confs_out, const Matrix_i32 markers, const Weka_model_tree tree, const char *weka_jar_fname)
	Predicts the labels for a set of marker samples using a Weka J48 model tree.
Error *	predict_labels_with_weka_part_model_tree (Vector_u32 labels_out, Vector_d confs_out, const Matrix_i32 markers, const Weka_model_tree tree, const char *weka_jar_fname)
	Predicts the labels for a set of marker samples using a Weka PART model tree.
Error *	read_weka_model_tree (Weka_model_tree *tree_out, const char tree_xml_fname, const char tree_dtd_fname, const char model_dirname)
	Reads a Weka model tree.
void	free_weka_model_tree (Weka_model_tree *tree)
	Frees a Weka model tree.

Detailed Description

Definitions for a Weka J48 and PART classifiers.

Author:: Joseph Schlecht

License:: Creative Commons BY-NC-SA 3.0

Definition in file weka.c.

Function Documentation

static Error* read_weka_model_labels	(	char **	labels_str_out,
		const char *	labels_fname
	)		`[static]`

Reads a list of Weka model labels.

Definition at line 76 of file weka.c.

Error* train_weka_j48_model	(	const Vector_u32 *	labels,
		const Matrix_i32 *	markers,
		const char *	labels_fname,
		const char *	model_fname,
		const char *	weka_jar_fname
	)

Trains a Weka J48 model and writes it to a file.

Parameters:

labels	Labels for training, with ith element as corresponding to the ith sample in markers.
markers	Markers for training, with ith row as a sample corresponding to the ith label in labels.
labels_fname	File name containing labels for groups in the model.
model_fname	File name containing the model.
weka_jar_fname	Weka java archive file.

Returns:: On success, NULL is returned; otherwise an error is returned.

Definition at line 141 of file weka.c.

Error* train_weka_part_model	(	const Vector_u32 *	labels,
		const Matrix_i32 *	markers,
		const char *	labels_fname,
		const char *	model_fname,
		const char *	weka_jar_fname
	)

Trains a Weka PART model and writes it to a file.

Parameters:

labels	Labels for training, with ith element as corresponding to the ith sample in markers.
markers	Markers for training, with ith row as a sample corresponding to the ith label in labels.
labels_fname	File name containing labels for groups in the model.
model_fname	File name containing the model.
weka_jar_fname	Weka java archive file.

Returns:: On success, NULL is returned; otherwise an error is returned.

Definition at line 232 of file weka.c.

Error* predict_labels_with_weka_j48_model	(	Vector_u32 **	labels_out,
		Vector_d **	confs_out,
		const Matrix_i32 *	markers,
		const char *	labels_fname,
		const char *	model_fname,
		const char *	weka_jar_fname
	)

Predicts the labels for a set of marker samples using Weka's J48 classifier.

Parameters:

labels_out	Result parameter.
confs_out	Result parameter.
markers	Marker data to predict.
labels_fname	File name containing labels used by the model.
model_fname	File name containing the model.
weka_jar_fname	Weka java archive file.

Returns:: On success, NULL is returned; otherwise an error is returned.

Definition at line 323 of file weka.c.

Error* predict_labels_with_weka_part_model	(	Vector_u32 **	labels_out,
		Vector_d **	confs_out,
		const Matrix_i32 *	markers,
		const char *	labels_fname,
		const char *	model_fname,
		const char *	weka_jar_fname
	)

Predicts the labels for a set of marker samples using Weka's PART classifier.

Parameters:

labels_out	Result parameter.
confs_out	Result parameter.
markers	Marker data to predict.
labels_fname	File name containing labels used by the model.
model_fname	File name containing the model.
weka_jar_fname	Weka java archive file.

Returns:: On success, NULL is returned; otherwise an error is returned.

Definition at line 487 of file weka.c.

static Error* read_weka_xml_doc	(	xmlDoc **	xml_doc_out,
		const char *	xml_fname,
		const char *	dtd_fname
	)		`[static]`

Reads and and optionally validates an XML document.

Definition at line 642 of file weka.c.

static Error* create_weka_model_tree_from_xml_node	(	Weka_model_tree **	tree_out,
		const Weka_model_tree *	parent,
		uint32_t	parent_label,
		xmlNode *	xml_node,
		const char *	model_dirname
	)		`[static]`

Allocates and initializes a model tree from an xml node.

Definition at line 690 of file weka.c.

static Error* create_model_training_data	(	Vector_u32 **	train_labels_out,
		Matrix_i32 **	train_markers_out,
		const Vector_u32 *	data_labels,
		const Matrix_i32 *	data_markers,
		const Vector_u32 *	model_labels,
		Vector_u32 const	model_altlabels
	)		`[static]`

Creates a set of training data as a model-specific labeled subset of a larger data set.

Definition at line 825 of file weka.c.

static Error* train_weka_j48_model_node	(	Weka_model_node *	node,
		const Vector_u32 *	labels,
		const Matrix_i32 *	markers,
		const char *	weka_jar_fname
	)		`[static]`

Trains a model in the tree using node-specific data.

Definition at line 896 of file weka.c.

static Error* train_weka_part_model_node	(	Weka_model_node *	node,
		const Vector_u32 *	labels,
		const Matrix_i32 *	markers,
		const char *	weka_jar_fname
	)		`[static]`

Trains a model in the tree using node-specific data.

Definition at line 944 of file weka.c.

Error* train_weka_j48_model_tree	(	Weka_model_tree **	tree_out,
		const Vector_u32 *	labels,
		const Matrix_i32 *	markers,
		const char *	tree_xml_fname,
		const char *	tree_dtd_fname,
		const char *	model_dirname,
		const char *	weka_jar_fname
	)

Trains a Weka J48 model tree .

Parameters:

tree_out	Result parameter.
labels	Sample group labels.
markers	Sample marker values.
tree_xml_fname	XML file containing the model tree information.
tree_dtd_fname	DTD file for validating the XML file, can be NULL.
model_dirname	Model directory location.
weka_jar_fname	Weka java archive file.

Definition at line 1000 of file weka.c.

Error* train_weka_part_model_tree	(	Weka_model_tree **	tree_out,
		const Vector_u32 *	labels,
		const Matrix_i32 *	markers,
		const char *	tree_xml_fname,
		const char *	tree_dtd_fname,
		const char *	model_dirname,
		const char *	weka_jar_fname
	)

Trains a Weka PART model tree .

Parameters:

tree_out	Result parameter.
labels	Sample group labels.
markers	Sample marker values.
tree_xml_fname	XML file containing the model tree information.
tree_dtd_fname	DTD file for validating the XML file, can be NULL.
model_dirname	Model directory location.
weka_jar_fname	Weka java archive file.

Definition at line 1061 of file weka.c.

static Error* recursively_predict_j48_labels_in_model_tree	(	Vector_u32 **	labels_out,
		Vector_d **	confs_in_out,
		const Weka_model_tree *	tree,
		const Matrix_i32 *	markers,
		const char *	weka_jar_fname
	)		`[static]`

Recursively predicts a label from a model tree.

Definition at line 1114 of file weka.c.

static Error* recursively_predict_part_labels_in_model_tree	(	Vector_u32 **	labels_out,
		Vector_d **	confs_in_out,
		const Weka_model_tree *	tree,
		const Matrix_i32 *	markers,
		const char *	weka_jar_fname
	)		`[static]`

Recursively predicts a label from a model tree.

Definition at line 1208 of file weka.c.

Error* predict_labels_with_weka_j48_model_tree	(	Vector_u32 **	labels_out,
		Vector_d **	confs_out,
		const Matrix_i32 *	markers,
		const Weka_model_tree *	tree,
		const char *	weka_jar_fname
	)

Predicts the labels for a set of marker samples using a Weka J48 model tree.

Parameters:

labels_out	Result parameter. If *labels_out is NULL, it is allocated; otherwise its space is re-used.
confs_out	Result parameter. If *confs_out is NULL, it is allocated; otherwise its space is re-used.
markers	Marker data to predict. Each row is a sample for prediction, corresponding to an element in the result parameters.
tree	Trained model tree to use for predicting.
weka_jar_fname	Weka java archive file.

Returns:: On success, NULL is returned; otherwise an error is returned, but the result parameters are not freed.

Definition at line 1315 of file weka.c.

Error* predict_labels_with_weka_part_model_tree	(	Vector_u32 **	labels_out,
		Vector_d **	confs_out,
		const Matrix_i32 *	markers,
		const Weka_model_tree *	tree,
		const char *	weka_jar_fname
	)

Predicts the labels for a set of marker samples using a Weka PART model tree.

Parameters:

labels_out	Result parameter. If *labels_out is NULL, it is allocated; otherwise its space is re-used.
confs_out	Result parameter. If *confs_out is NULL, it is allocated; otherwise its space is re-used.
markers	Marker data to predict. Each row is a sample for prediction, corresponding to an element in the result parameters.
tree	Trained model tree to use for predicting.
weka_jar_fname	Weka java archive file.

Returns:: On success, NULL is returned; otherwise an error is returned, but the result parameters are not freed.

Definition at line 1352 of file weka.c.

Error* read_weka_model_tree	(	Weka_model_tree **	tree_out,
		const char *	tree_xml_fname,
		const char *	tree_dtd_fname,
		const char *	model_dirname
	)

Reads a Weka model tree.

Parameters:

tree_out	Result parameter.
tree_xml_fname	XML file containing the model tree information.
tree_dtd_fname	DTD file for validating the XML file, can be NULL.
model_dirname	Model directory location.

Definition at line 1381 of file weka.c.

void free_weka_model_tree ( Weka_model_tree * tree )

Frees a Weka model tree.

Parameters:

tree	Model tree to free.

Definition at line 1423 of file weka.c.

Functions

Detailed Description

Function Documentation