Merge pull request #2 from drkrillo/sequential_minimum_optimization_a…

…dd_doctests

Sequential minimum optimization add doctests

machine_learning/sequential_minimum_optimization.py

@@ -395,8 +395,8 @@ def _get_new_alpha(self, i1, i2, a1, a2, e1, e2, y1, y2):
     # Normalize data using min-max method
     def _norm(self, data):
         if self._init:
-            self._min = np.min(data, axis=0) # [0, -2, 2] [1, 1, 0]
-            self._max = np.max(data, axis=0) # [1, -1, 3] [1, 1, 1]
+            self._min = np.min(data, axis=0)
+            self._max = np.max(data, axis=0)
             self._init = False
             return (data - self._min) / (self._max - self._min)
         else:
@@ -436,6 +436,12 @@ def length(self):
 
 
 class Kernel:
+    """
+    >>> from machine_learning.sequential_minimum_optimization import Kernel
+    >>> kernel = Kernel(kernel='linear')
+    >>> kernel._kernel_name
+    'linear'
+    """
     def __init__(self, kernel, degree=1.0, coef0=0.0, gamma=1.0):
         self.degree = np.float64(degree)
         self.coef0 = np.float64(coef0)
@@ -445,9 +451,23 @@ def __init__(self, kernel, degree=1.0, coef0=0.0, gamma=1.0):
         self._check()
 
     def _polynomial(self, v1, v2):
+        """
+        >>> from machine_learning.sequential_minimum_optimization import Kernel
+        >>> kernel = Kernel(kernel='linear')
+        >>> result = kernel._polynomial(np.array([1, 2]), np.array([2, 3]))
+        >>> int(result) == 8
+        True
+        """
         return (self.gamma * np.inner(v1, v2) + self.coef0) ** self.degree
 
     def _linear(self, v1, v2):
+        """
+        >>> from machine_learning.sequential_minimum_optimization import Kernel
+        >>> kernel = Kernel(kernel='linear')
+        >>> result = kernel._polynomial(np.array([1, 2]), np.array([2, 3]))
+        >>> int(result) == 8
+        True
+        """
         return np.inner(v1, v2) + self.coef0
 
     def _rbf(self, v1, v2):