liblinear-ruby 0.0.4 → 0.0.5

data/{liblinear-1.94 → liblinear-1.95}/predict.c

@@ -158,9 +158,7 @@ void do_predict(FILE *input, FILE *output)
 		sumpt += predict_label*target_label;
 		++total;
 	}
-	if(model_->param.solver_type==L2R_L2LOSS_SVR ||
-	   model_->param.solver_type==L2R_L1LOSS_SVR_DUAL ||
-	   model_->param.solver_type==L2R_L2LOSS_SVR_DUAL)
+	if(check_regression_model(model_))
 	{
 		info("Mean squared error = %g (regression)\n",error/total);
 		info("Squared correlation coefficient = %g (regression)\n",

data/{liblinear-1.94 → liblinear-1.95}/python/README

@@ -205,6 +205,36 @@ LIBLINEAR shared library:
     >>> nr_class = model_.get_nr_class()
     >>> class_labels = model_.get_labels()
     >>> is_prob_model = model_.is_probability_model()
+    >>> is_regression_model = model_.is_regression_model()
+
+    The decision function is W*x + b, where
+        W is an nr_class-by-nr_feature matrix, and
+        b is a vector of size nr_class.
+    To access W_kj (i.e., coefficient for the k-th class and the j-th feature)
+    and b_k (i.e., bias for the k-th class), use the following functions.
+
+    >>> W_kj = model_.get_decfun_coef(feat_idx=j, label_idx=k)
+    >>> b_k = model_.get_decfun_bias(label_idx=k)
+
+    We also provide a function to extract w_k (i.e., the k-th row of W) and
+    b_k directly as follows.
+
+    >>> [w_k, b_k] = model_.get_decfun(label_idx=k)
+
+    Note that w_k is a Python list of length nr_feature, which means that
+        w_k[0] = W_k1.
+    For regression models, W is just a vector of length nr_feature. Either
+	set label_idx=0 or omit the label_idx parameter to access the coefficients.
+
+    >>> W_j = model_.get_decfun_coef(feat_idx=j)
+    >>> b = model_.get_decfun_bias()
+    >>> [W, b] = model_.get_decfun()
+
+    Note that in get_decfun_coef, get_decfun_bias, and get_decfun, feat_idx
+    starts from 1, while label_idx starts from 0. If label_idx is not in the
+    valid range (0 to nr_class-1), then a NaN will be returned; and if feat_idx
+    is not in the valid range (1 to nr_feature), then a zero value will be
+    returned. For regression models, label_idx is ignored.
 
 Utility Functions
 =================

data/{liblinear-1.94 → liblinear-1.95}/python/liblinear.py

@@ -5,12 +5,18 @@ from ctypes.util import find_library
 from os import path
 import sys
 
+__all__ = ['liblinear', 'feature_node', 'gen_feature_nodearray', 'problem',
+           'parameter', 'model', 'toPyModel', 'L2R_LR', 'L2R_L2LOSS_SVC_DUAL',
+           'L2R_L2LOSS_SVC', 'L2R_L1LOSS_SVC_DUAL', 'MCSVM_CS', 
+           'L1R_L2LOSS_SVC', 'L1R_LR', 'L2R_LR_DUAL', 'L2R_L2LOSS_SVR', 
+           'L2R_L2LOSS_SVR_DUAL', 'L2R_L1LOSS_SVR_DUAL', 'print_null']
+
 try:
 	dirname = path.dirname(path.abspath(__file__))
 	if sys.platform == 'win32':
 		liblinear = CDLL(path.join(dirname, r'..\windows\liblinear.dll'))
 	else:
-		liblinear = CDLL(path.join(dirname, '../liblinear.so.1'))
+		liblinear = CDLL(path.join(dirname, '../liblinear.so.2'))
 except:
 # For unix the prefix 'lib' is not considered.
 	if find_library('linear'):
@@ -20,13 +26,17 @@ except:
 	else:
 		raise Exception('LIBLINEAR library not found.')
 
-# Construct constants
-SOLVER_TYPE = ['L2R_LR', 'L2R_L2LOSS_SVC_DUAL', 'L2R_L2LOSS_SVC', 'L2R_L1LOSS_SVC_DUAL',\
-		'MCSVM_CS', 'L1R_L2LOSS_SVC', 'L1R_LR', 'L2R_LR_DUAL', \
-		None, None, None, \
-		'L2R_L2LOSS_SVR', 'L2R_L2LOSS_SVR_DUAL', 'L2R_L1LOSS_SVR_DUAL']
-for i, s in enumerate(SOLVER_TYPE): 
-	if s is not None: exec("%s = %d" % (s , i))
+L2R_LR = 0
+L2R_L2LOSS_SVC_DUAL = 1 
+L2R_L2LOSS_SVC = 2 
+L2R_L1LOSS_SVC_DUAL = 3
+MCSVM_CS = 4 
+L1R_L2LOSS_SVC = 5 
+L1R_LR = 6 
+L2R_LR_DUAL = 7  
+L2R_L2LOSS_SVR = 11
+L2R_L2LOSS_SVR_DUAL = 12
+L2R_L1LOSS_SVR_DUAL = 13
 
 PRINT_STRING_FUN = CFUNCTYPE(None, c_char_p)
 def print_null(s): 
@@ -240,9 +250,23 @@ class model(Structure):
 		liblinear.get_labels(self, labels)
 		return labels[:nr_class]
 
+	def get_decfun_coef(self, feat_idx, label_idx=0):
+		return liblinear.get_decfun_coef(self, feat_idx, label_idx)
+
+	def get_decfun_bias(self, label_idx=0):
+		return liblinear.get_decfun_bias(self, label_idx)
+
+	def get_decfun(self, label_idx=0):
+		w = [liblinear.get_decfun_coef(self, feat_idx, label_idx) for feat_idx in range(1, self.nr_feature+1)]
+		b = liblinear.get_decfun_bias(self, label_idx)
+		return (w, b)
+
 	def is_probability_model(self):
 		return (liblinear.check_probability_model(self) == 1)
 
+	def is_regression_model(self):
+		return (liblinear.check_regression_model(self) == 1)
+
 def toPyModel(model_ptr):
 	"""
 	toPyModel(model_ptr) -> model
@@ -268,10 +292,13 @@ fillprototype(liblinear.load_model, POINTER(model), [c_char_p])
 fillprototype(liblinear.get_nr_feature, c_int, [POINTER(model)])
 fillprototype(liblinear.get_nr_class, c_int, [POINTER(model)])
 fillprototype(liblinear.get_labels, None, [POINTER(model), POINTER(c_int)])
+fillprototype(liblinear.get_decfun_coef, c_double, [POINTER(model), c_int, c_int])
+fillprototype(liblinear.get_decfun_bias, c_double, [POINTER(model), c_int])
 
 fillprototype(liblinear.free_model_content, None, [POINTER(model)])
 fillprototype(liblinear.free_and_destroy_model, None, [POINTER(POINTER(model))])
 fillprototype(liblinear.destroy_param, None, [POINTER(parameter)])
 fillprototype(liblinear.check_parameter, c_char_p, [POINTER(problem), POINTER(parameter)])
 fillprototype(liblinear.check_probability_model, c_int, [POINTER(model)])
+fillprototype(liblinear.check_regression_model, c_int, [POINTER(model)])
 fillprototype(liblinear.set_print_string_function, None, [CFUNCTYPE(None, c_char_p)])

data/{liblinear-1.94 → liblinear-1.95}/python/liblinearutil.py

@@ -3,6 +3,12 @@
 import os, sys
 sys.path = [os.path.dirname(os.path.abspath(__file__))] + sys.path 
 from liblinear import *
+from liblinear import __all__ as liblinear_all
+from ctypes import c_double
+
+__all__ = ['svm_read_problem', 'load_model', 'save_model', 'evaluations',
+           'train', 'predict'] + liblinear_all
+
 
 def svm_read_problem(data_file_name):
 	"""
@@ -241,7 +247,7 @@ def predict(y, x, m, options=""):
 		y = [0] * len(x)
 	ACC, MSE, SCC = evaluations(y, pred_labels)
 	l = len(y)
-	if solver_type in [L2R_L2LOSS_SVR, L2R_L2LOSS_SVR_DUAL, L2R_L1LOSS_SVR_DUAL]:
+	if m.is_regression_model():
 		info("Mean squared error = %g (regression)" % MSE)
 		info("Squared correlation coefficient = %g (regression)" % SCC)
 	else:

data/{liblinear-1.94 → liblinear-1.95}/train.c

@@ -300,7 +300,7 @@ void parse_command_line(int argc, char **argv, char *input_file_name, char *mode
 void read_problem(const char *filename)
 {
 	int max_index, inst_max_index, i;
-	long int elements, j;
+	size_t elements, j;
 	FILE *fp = fopen(filename,"r");
 	char *endptr;
 	char *idx, *val, *label;

data/spec/liblinear/model_spec.rb

@@ -4,8 +4,11 @@ require 'liblinear'
 describe Liblinear::Model do
   before do
     @prob = Liblinear::Problem.new([1, 2], [[1],[2]])
-    @param = Liblinear::Parameter.new
-    @model = Liblinear::Model.new(@prob, @param)
+    @param_1 = Liblinear::Parameter.new
+    @model_classification = Liblinear::Model.new(@prob, @param_1)
+
+    @param_2 = Liblinear::Parameter.new({ solver_type: Liblinear::L2R_L2LOSS_SVR })
+    @model_regression = Liblinear::Model.new(@prob, @param_2)
   end
 
   describe '#initialize' do
@@ -30,33 +33,69 @@ describe Liblinear::Model do
     end
   end
 
-  describe '#nr_class' do
+  describe '#class_size' do
     it 'returns the number of classes' do
-      expect(@model.nr_class).to eq(2)
+      expect(@model_classification.class_size).to eq(2)
+    end
+  end
+
+  describe '#feature_size' do
+    it 'returns the number of features' do
+      expect(@model_classification.feature_size).to eq(1)
     end
   end
 
   describe '#labels' do
     it 'returns labels' do
-      expect(@model.labels).to eq([1, 2])
+      expect(@model_classification.labels).to eq([1, 2])
     end
   end
 
   describe '#predict' do
-    it 'return predicted class' do
-      expect(@model.predict([1]).class).to eq(Float)
+    it 'returns predicted class' do
+      expect(@model_classification.predict([1]).class).to eq(Float)
     end
   end
 
   describe '#predict_probability' do
-    it 'return predict_probability' do
-      expect(@model.predict_probability([1]).class).to eq(Hash)
+    it 'returns predict_probability' do
+      expect(@model_classification.predict_probability([1]).class).to eq(Hash)
     end
   end
 
   describe '#predict_values' do
-    it 'return predict_values' do
-      expect(@model.predict_values([1]).class).to eq(Hash)
+    it 'returns predict_values' do
+      expect(@model_classification.predict_values([1]).class).to eq(Hash)
+    end
+  end
+
+  describe '#coefficient' do
+    it 'returns a coefficient' do
+      expect(@model_classification.coefficient(1, 0).class).to eq(Float)
+    end
+
+    it 'returns a coefficient' do
+      expect(@model_classification.coefficient(1).class).to eq(Float)
+    end
+
+    it 'returns all coefficients' do
+      expect(@model_classification.coefficient.class).to eq(Array)
+    end
+  end
+
+  describe '#bias' do
+    it 'return a bias' do
+      expect(@model_classification.bias.class).to eq(Float)
+    end
+  end
+
+  describe '#regression_model?' do
+    it 'returns true' do
+      expect(@model_regression.regression_model?).to eq(true)
+    end
+
+    it 'returns false' do
+      expect(@model_classification.regression_model?).to eq (false)
     end
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: liblinear-ruby
 version: !ruby/object:Gem::Version
-  version: 0.0.4
+  version: 0.0.5
 platform: ruby
 authors:
 - Kei Tsuchiya
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-08-07 00:00:00.000000000 Z
+date: 2014-11-03 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -70,45 +70,45 @@ files:
 - lib/liblinear/parameter.rb
 - lib/liblinear/problem.rb
 - lib/liblinear/version.rb
-- liblinear-1.94/COPYRIGHT
-- liblinear-1.94/Makefile
-- liblinear-1.94/Makefile.win
-- liblinear-1.94/README
-- liblinear-1.94/blas/Makefile
-- liblinear-1.94/blas/blas.h
-- liblinear-1.94/blas/blasp.h
-- liblinear-1.94/blas/daxpy.c
-- liblinear-1.94/blas/ddot.c
-- liblinear-1.94/blas/dnrm2.c
-- liblinear-1.94/blas/dscal.c
-- liblinear-1.94/heart_scale
-- liblinear-1.94/linear.cpp
-- liblinear-1.94/linear.def
-- liblinear-1.94/linear.h
-- liblinear-1.94/matlab/Makefile
-- liblinear-1.94/matlab/README
-- liblinear-1.94/matlab/libsvmread.c
-- liblinear-1.94/matlab/libsvmwrite.c
-- liblinear-1.94/matlab/linear_model_matlab.c
-- liblinear-1.94/matlab/linear_model_matlab.h
-- liblinear-1.94/matlab/make.m
-- liblinear-1.94/matlab/predict.c
-- liblinear-1.94/matlab/train.c
-- liblinear-1.94/predict.c
-- liblinear-1.94/python/Makefile
-- liblinear-1.94/python/README
-- liblinear-1.94/python/liblinear.py
-- liblinear-1.94/python/liblinearutil.py
-- liblinear-1.94/train.c
-- liblinear-1.94/tron.cpp
-- liblinear-1.94/tron.h
-- liblinear-1.94/windows/liblinear.dll
-- liblinear-1.94/windows/libsvmread.mexw64
-- liblinear-1.94/windows/libsvmwrite.mexw64
-- liblinear-1.94/windows/predict.exe
-- liblinear-1.94/windows/predict.mexw64
-- liblinear-1.94/windows/train.exe
-- liblinear-1.94/windows/train.mexw64
+- liblinear-1.95/COPYRIGHT
+- liblinear-1.95/Makefile
+- liblinear-1.95/Makefile.win
+- liblinear-1.95/README
+- liblinear-1.95/blas/Makefile
+- liblinear-1.95/blas/blas.h
+- liblinear-1.95/blas/blasp.h
+- liblinear-1.95/blas/daxpy.c
+- liblinear-1.95/blas/ddot.c
+- liblinear-1.95/blas/dnrm2.c
+- liblinear-1.95/blas/dscal.c
+- liblinear-1.95/heart_scale
+- liblinear-1.95/linear.cpp
+- liblinear-1.95/linear.def
+- liblinear-1.95/linear.h
+- liblinear-1.95/matlab/Makefile
+- liblinear-1.95/matlab/README
+- liblinear-1.95/matlab/libsvmread.c
+- liblinear-1.95/matlab/libsvmwrite.c
+- liblinear-1.95/matlab/linear_model_matlab.c
+- liblinear-1.95/matlab/linear_model_matlab.h
+- liblinear-1.95/matlab/make.m
+- liblinear-1.95/matlab/predict.c
+- liblinear-1.95/matlab/train.c
+- liblinear-1.95/predict.c
+- liblinear-1.95/python/Makefile
+- liblinear-1.95/python/README
+- liblinear-1.95/python/liblinear.py
+- liblinear-1.95/python/liblinearutil.py
+- liblinear-1.95/train.c
+- liblinear-1.95/tron.cpp
+- liblinear-1.95/tron.h
+- liblinear-1.95/windows/liblinear.dll
+- liblinear-1.95/windows/libsvmread.mexw64
+- liblinear-1.95/windows/libsvmwrite.mexw64
+- liblinear-1.95/windows/predict.exe
+- liblinear-1.95/windows/predict.mexw64
+- liblinear-1.95/windows/train.exe
+- liblinear-1.95/windows/train.mexw64
 - liblinear-ruby.gemspec
 - spec/liblinear/model_spec.rb
 - spec/liblinear/parameter_spec.rb
@@ -134,7 +134,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.0.14
+rubygems_version: 2.0.2
 signing_key: 
 specification_version: 4
 summary: Ruby wrapper of LIBLINEAR using SWIG
@@ -143,3 +143,4 @@ test_files:
 - spec/liblinear/parameter_spec.rb
 - spec/liblinear/problem_spec.rb
 - spec/liblinear_spec.rb
+has_rdoc: