Add 4 Normalizing methods for data preprocessing and update Readme.md

README.md

@@ -22,6 +22,7 @@ but rather to present the inner workings of them in a transparent and accessible
     + [Genetic Algorithm](#genetic-algorithm)
     + [Association Analysis](#association-analysis)
   * [Implementations](#implementations)
+    + [Normalizing](#normalizing)
     + [Supervised Learning](#supervised-learning)
     + [Unsupervised Learning](#unsupervised-learning)
     + [Reinforcement Learning](#reinforcement-learning)
@@ -267,6 +268,11 @@ but rather to present the inner workings of them in a transparent and accessible
 
 
 ## Implementations
+### Normalizing
+- [MinMaxScaler](mlfromscratch/normalizing/minmaxscaler.py)
+- [StandardScaler](mlfromscratch/normalizing/standardscaler.py)
+- [StandardScaler With Selective mean and variance](mlfromscratch/normalizing/standardscalerwithmv.py)
+- [MaxAbsScaler](mlfromscratch/normalizing/maxabsscaler.py)
 ### Supervised Learning
 - [Adaboost](mlfromscratch/supervised_learning/adaboost.py)
 - [Bayesian Regression](mlfromscratch/supervised_learning/bayesian_regression.py)

mlfromscratch/normalizing/__init__.py

@@ -0,0 +1,4 @@
+from .minmaxscaler import MinMaxScaler
+from .standardscaler import StandardScaler
+from .standardscalerwithmv import StandardScalerAdv
+from .maxabsscaler import MaxAbsScaler

mlfromscratch/normalizing/maxabsscaler.py

@@ -0,0 +1,11 @@
+from __future__ import print_function, division
+import numpy as np
+
+def MaxAbsScaler(x):
+    """
+    Parameters :
+    x : Dataset
+
+    """
+    normalized = x/np.max(abs(x))
+    return normalized

mlfromscratch/normalizing/minmaxscaler.py

@@ -0,0 +1,13 @@
+from __future__ import print_function, division
+import numpy as np
+
+def MinMaxScaler(x,minmax):
+    """
+    Parameters :
+    x : Dataset
+    minmax : the minimum and maximum range of your features
+
+    """
+    normalized = (x-np.min(x))/(np.max(x)-np.min(x))
+    scale = (normalized * (np.max(minmax)-np.min(minmax))) + np.min(minmax)
+    return scale

mlfromscratch/normalizing/standardscaler.py

@@ -0,0 +1,16 @@
+from __future__ import print_function, division
+import numpy as np
+
+def StandardScaler(x,standard=True):
+    """
+    Parameters :
+    x : Dataset
+    standard : True or False (True = setting the mean of 0 and var of 1, False = computing mean and var of data)
+
+    """
+    if standard == True:
+        normalized = (x-0)/np.sqrt(1)
+
+    else:
+        normalized = (x-np.mean(x))/np.std(x)
+    return normalized

mlfromscratch/normalizing/standardscalerwithmv.py

@@ -0,0 +1,12 @@
+from __future__ import print_function, division
+import numpy as np
+
+def StandardScalerAdv(x,mean,var):
+    """
+    Parameters :
+    x : Dataset
+    mean : your favorite mean for scaling features
+    var : your favorite variance for scaling features
+
+    """
+    normalized = (x-mean)/np.sqrt(var)