Modify Python interface, major changes are:

    + Find local library first.
    + Change show() to __str__.

diff --git a/python/README b/python/README
index 40fe87d1e2ffc0e81f0bf9f8c06baef8589c4507..1c7094cf7add37828e7b0aa568cbcbce771eaede 100644 (file)
--- a/python/README
+++ b/python/README
@@ -118,6 +118,10 @@ LIBLINEAR shared library:
 
     val: a float indicates the feature value.
 
+    Show the index and the value of a node.
+
+    >>> print(node) 
+
 - Function: gen_feature_nodearray(xi [,feature_max=None [,issparse=True]])
 
     Generate a feature vector from a Python list/tuple or a dictionary:
@@ -173,7 +177,7 @@ LIBLINEAR shared library:
 
     Show values of parameters.
 
-    >>> param.show()
+    >>> print(param)
 
 - class model:
 

diff --git a/python/liblinear.py b/python/liblinear.py
index 506f118bd01f5f96d47567f6112ff12039e97fd6..8db2b88bb1d5bcb1c6b7bc6f80dbabdf00b9c484 100644 (file)
--- a/python/liblinear.py
+++ b/python/liblinear.py
@@ -2,21 +2,23 @@
 
 from ctypes import *
 from ctypes.util import find_library
+from os import path
 import sys
-import os
 
-# For unix the prefix 'lib' is not considered.
-if find_library('linear'):
-       liblinear = CDLL(find_library('linear'))
-elif find_library('liblinear'):
-       liblinear = CDLL(find_library('liblinear'))
-else:
+try:
+       dirname = path.dirname(path.abspath(__file__))
        if sys.platform == 'win32':
-               liblinear = CDLL(os.path.join(os.path.dirname(__file__),\
-                               '../windows/liblinear.dll'))
+               liblinear = CDLL(path.join(dirname, r'..\windows\liblinear.dll'))
+       else:
+               liblinear = CDLL(path.join(dirname, '../liblinear.so.1'))
+except:
+# For unix the prefix 'lib' is not considered.
+       if find_library('linear'):
+               liblinear = CDLL(find_library('linear'))
+       elif find_library('liblinear'):
+               liblinear = CDLL(find_library('liblinear'))
        else:
-               liblinear = CDLL(os.path.join(os.path.dirname(__file__),\
-                               '../liblinear.so.1'))
+               raise Exception('LIBLINEAR library not found.')
 
 # Construct constants
 SOLVER_TYPE = ['L2R_LR', 'L2R_L2LOSS_SVC_DUAL', 'L2R_L2LOSS_SVC', 'L2R_L1LOSS_SVC_DUAL',\
@@ -42,6 +44,9 @@ class feature_node(Structure):
        _types = [c_int, c_double]
        _fields_ = genFields(_names, _types)
 
+       def __str__(self):
+               return '%d:%g' % (self.index, self.value)
+
 def gen_feature_nodearray(xi, feature_max=None, issparse=True):
        if isinstance(xi, dict):
                index_range = xi.keys()
@@ -121,11 +126,15 @@ class parameter(Structure):
                        options = ''
                self.parse_options(options)
 
-       def show(self):
-               attrs = parameter._names + self.__dict__.keys()
+       def __str__(self):
+               s = ''
+               attrs = parameter._names + list(self.__dict__.keys())
                values = map(lambda attr: getattr(self, attr), attrs) 
                for attr, val in zip(attrs, values):
-                       print(' %s: %s' % (attr, val))
+                       s += (' %s: %s\n' % (attr, val))
+               s = s.strip()
+
+               return s
 
        def set_to_default_values(self):
                self.solver_type = L2R_L2LOSS_SVC_DUAL
@@ -141,7 +150,12 @@ class parameter(Structure):
                self.print_func = None
 
        def parse_options(self, options):
-               argv = options.split()
+               if isinstance(options, list):
+                       argv = options
+               elif isinstance(options, str):
+                       argv = options.split()
+               else:
+                       raise TypeError("arg 1 should be a list or a str.")
                self.set_to_default_values()
                self.print_func = cast(None, PRINT_STRING_FUN)
                weight_label = []
@@ -261,5 +275,3 @@ 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.set_print_string_function, None, [CFUNCTYPE(None, c_char_p)])
-
-

diff --git a/python/liblinearutil.py b/python/liblinearutil.py
index 1940abe7351fb61a6e444e17d963f868f98f4e79..d63e088df16099d928c023f14ba275c7d60cf13d 100644 (file)
--- a/python/liblinearutil.py
+++ b/python/liblinearutil.py
@@ -1,5 +1,7 @@
 #!/usr/bin/env python
 
+import os, sys
+sys.path = [os.path.dirname(os.path.abspath(__file__))] + sys.path 
 from liblinear import *
 
 def svm_read_problem(data_file_name):
@@ -76,8 +78,8 @@ def evaluations(ty, pv):
 
 def train(arg1, arg2=None, arg3=None):
        """
-       train(y, x [, 'options']) -> model | ACC
-       train(prob, [, 'options']) -> model | ACC
+       train(y, x [, options]) -> model | ACC
+       train(prob [, options]) -> model | ACC
        train(prob, param) -> model | ACC
 
        Train a model from data (y, x) or a problem prob using
@@ -85,7 +87,7 @@ def train(arg1, arg2=None, arg3=None):
        If '-v' is specified in 'options' (i.e., cross validation)
        either accuracy (ACC) or mean-squared error (MSE) is returned.
 
-       'options':
+       options:
                -s type : set type of solver (default 1)
                  for multi-class classification
                         0 -- L2-regularized logistic regression (primal)
@@ -158,16 +160,14 @@ def train(arg1, arg2=None, arg3=None):
                m = liblinear.train(prob, param)
                m = toPyModel(m)
 
-               # If prob is destroyed, data including SVs pointed by m can remain.
-               m.x_space = prob.x_space
                return m
 
 def predict(y, x, m, options=""):
        """
-       predict(y, x, m [, "options"]) -> (p_labels, p_acc, p_vals)
+       predict(y, x, m [, options]) -> (p_labels, p_acc, p_vals)
 
        Predict data (y, x) with the SVM model m.
-       "options":
+       options:
            -b probability_estimates: whether to output probability estimates, 0 or 1 (default 0); currently for logistic regression only
            -q quiet mode (no outputs)