fix some typos

jubatus · Apr 17, 2017 · 9c3f45d · 9c3f45d
1 parent 722a772
commit 9c3f45d
Show file tree

Hide file tree

Showing 5 changed files with 29 additions and 25 deletions.
diff --git a/example/regression_csv.py b/example/regression_csv.py
@@ -4,10 +4,14 @@
 from __future__ import absolute_import, division, print_function, unicode_literals
 
 """
-Using Regression and Boston dataset
+Using Regression and CSV file
 ==================================================
 
-In this example we show regression using Boston dataset.
+This is a simple example that illustrates:
+
+* How to load CSV files and convert int into Jubakit dataset.
+* Training the regression using the dataset.
+* Getting regression result.
 """
 
 import numpy as np

diff --git a/example/regression_sklearn_wrapper.py b/example/regression_sklearn_wrapper.py
@@ -21,7 +21,7 @@
 from sklearn.pipeline import Pipeline
 from sklearn.utils import shuffle
 
-# load hand-writtern number recognition dataset
+# load boston housing values dataset
 boston = load_boston()
 # shuffle and separate the dataset
 X, y = shuffle(boston.data, boston.target, random_state=42)

diff --git a/jubakit/regression.py b/jubakit/regression.py
@@ -179,8 +179,8 @@ def _default_parameter(cls, method):
         return {'learning_rate': 1.0}
       elif method in ('PA', 'passive_aggressive'):
         return {'sensitivity': 1.0}
-      elif method in ('PA1', 'passive_aggressive1',
-                      'PA2', 'passive_aggressive2',
+      elif method in ('PA1', 'passive_aggressive_1',
+                      'PA2', 'passive_aggressive_2',
                       'CW', 'confidence_weighted',
                       'AROW',
                       'NHERD', 'normal_herd'):

diff --git a/jubakit/wrapper/classifier.py b/jubakit/wrapper/classifier.py
@@ -11,7 +11,7 @@ class BaseJubatusClassifier(BaseEstimator, ClassifierMixin):
   """
   scikit-learn Wrapper for Jubatus Classifiers.
   """
-  
+
   def __init__(self, n_iter=1, shuffle=False, softmax=False, embedded=True, seed=None):
     """
     Creates a base class for Jubatus Classifiers.
@@ -24,8 +24,8 @@ def __init__(self, n_iter=1, shuffle=False, softmax=False, embedded=True, seed=N
 
   @classmethod
   def _launch_classifier(self):
-    """ 
-    Launch Jubatus Classifier 
+    """
+    Launch Jubatus Classifier
     """
     raise NotImplementedError()
 
@@ -81,11 +81,11 @@ def decision_function(self, X):
 
   @classmethod
   def get_params(self, deep=True):
-    """ 
-    Return parameters. 
+    """
+    Return parameters.
     """
     raise NotImplementedError()
-  
+
   def set_params(self, **params):
     """
     Set parameters
@@ -95,7 +95,7 @@ def set_params(self, **params):
     return self
 
   def save(self, name):
-    """  
+    """
     Save the classifier model using name.
     """
     if self.classifier_ is not None:
@@ -111,22 +111,22 @@ def stop(self):
 
 
 class LinearClassifier(BaseJubatusClassifier):
-  
-  def __init__(self, method='AROW', regularization_weight=1.0, 
+
+  def __init__(self, method='AROW', regularization_weight=1.0,
                softmax=False, n_iter=1, shuffle=False, embedded=True, seed=None):
     super(LinearClassifier, self).__init__(n_iter, shuffle, softmax, embedded, seed)
     self.method = method
     self.regularization_weight = regularization_weight
-  
+
   def _launch_classifier(self):
     if self.method in ('perceptron', 'PA'):
       self.config_ = Config(method=self.method)
     elif self.method in ('PA1', 'PA2', 'CW', 'AROW', 'NHERD'):
-      self.config_ = Config(method=self.method, 
+      self.config_ = Config(method=self.method,
                             parameter={'regularization_weight': self.regularization_weight})
     else:
-      raise NotImplementedError('method {} is not implememented yet.'.format(self.method))
-    self.classifier_ = Classifier.run(config=self.config_, embedded=self.embedded) 
+      raise NotImplementedError('method {} is not implemented yet.'.format(self.method))
+    self.classifier_ = Classifier.run(config=self.config_, embedded=self.embedded)
 
   def get_params(self, deep=True):
     return {
@@ -141,7 +141,7 @@ def get_params(self, deep=True):
 
 
 class NearestNeighborsClassifier(BaseJubatusClassifier):
-  
+
   def __init__(self, method='euclid_lsh', nearest_neighbor_num=5, local_sensitivity=1.0,
                hash_num=128, n_iter=1, shuffle=False, softmax=False, embedded=True, seed=None):
     super(NearestNeighborsClassifier, self).__init__(n_iter, shuffle, softmax, embedded, seed)
@@ -157,12 +157,12 @@ def _launch_classifier(self):
                                                     'local_sensitivity': self.local_sensitivity,
                                                     'parameter': {'hash_num': self.hash_num}})
     elif self.method in ('euclidean', 'cosine'):
-      self.config_ = Config(method=self.method, 
+      self.config_ = Config(method=self.method,
                             parameter={'nearest_neighbor_num': self.nearest_neighbor_num,
                                        'local_sensitivity': self.local_sensitivity})
     else:
-      raise NotImplementedError('method {} is not implememented yet.'.format(self.method))                            
-    self.classifier_ = Classifier.run(config=self.config_, embedded=self.embedded) 
+      raise NotImplementedError('method {} is not implemented yet.'.format(self.method))
+    self.classifier_ = Classifier.run(config=self.config_, embedded=self.embedded)
 
   def get_params(self, deep=True):
     return {

diff --git a/jubakit/wrapper/regression.py b/jubakit/wrapper/regression.py
@@ -116,15 +116,15 @@ def _launch_regression(self):
     if self.method in ('perceptron'):
       self.config_ = Config(method=self.method,
                             parameter={'learning_rate': self.learning_rate})
-    if self.method in ('PA'):
+    elif self.method in ('PA'):
       self.config_ = Config(method=self.method,
                             parameter={'sensitivity': self.sensitivity})
     elif self.method in ('PA1', 'PA2', 'CW', 'AROW', 'NHERD'):
       self.config_ = Config(method=self.method,
                             parameter={'regularization_weight': self.regularization_weight,
                                        'sensitivity': self.sensitivity})
     else:
-      raise NotImplementedError('method {} is not implememented yet.'.format(self.method))
+      raise NotImplementedError('method {} is not implemented yet.'.format(self.method))
     self.regression_ = Regression.run(config=self.config_, embedded=self.embedded)
 
   def get_params(self, deep=True):
@@ -158,7 +158,7 @@ def _launch_regression(self):
       self.config_ = Config(method=self.method,
                             parameter={'nearest_neighbor_num': self.nearest_neighbor_num})
     else:
-      raise NotImplementedError('method {} is not implememented yet.'.format(self.method))
+      raise NotImplementedError('method {} is not implemented yet.'.format(self.method))
     self.regression_ = Regression.run(config=self.config_, embedded=self.embedded)
 
   def get_params(self, deep=True):