ENH More readable code

Author: luispedro

ch08/norm.py

@@ -2,40 +2,42 @@
 
 classNormalizePositive(object):
 
-deffit(self, X, y=None):
+deffit(self, features, y=None):
 # count features that are greater than zero in axis 0:
-binary= (X>0)
+binary= (features>0)
 count0=binary.sum(axis=0)
 
 # to avoid division by zero, set zero counts to one:
-count0+= (count0==0)
+count0[count0==0] =1.
 
-self.mean=X.sum(axis=0)/count0
+self.mean=features.sum(axis=0)/count0
 
 # Compute variance by average squared difference to the mean, but only
 # consider differences where binary is True (i.e., where there was a
 # true rating):
-diff= (X-self.mean) *binary
+diff= (features-self.mean) *binary
 diff**=2
+# regularize the estimate of std by adding 0.1
 self.std=np.sqrt(0.1+diff.sum(axis=0)/count0)
 returnself
 
-deffit_transform(self, X):
-returnself.fit(X).transform(X)
+deftransform(self, features):
+binary= (features>0)
+features=features-self.mean
+features/=self.std
+features*=binary
+returnfeatures
 
-deftransform(self, X):
-binary= (X>0)
-X=X-self.mean
-X/=self.std
-X*=binary
-returnX
-
-definverse_transform(self, X, copy=True):
+definverse_transform(self, features, copy=True):
 ifcopy:
-X=X.copy()
-X*=self.std
-X+=self.mean
-returnX
+features=features.copy()
+features*=self.std
+features+=self.mean
+returnfeatures
+
+deffit_transform(self, features):
+returnself.fit(features).transform(features)
+
 
 defpredict(train):
 norm=NormalizePositive()