In [3]:

    

%pylab inline


    

    




Populating the interactive namespace from numpy and matplotlib

In [30]:

    

# Import libraries
from __future__ import absolute_import, division, print_function

# Ignore warnings
import warnings
warnings.filterwarnings('ignore')

import sys
sys.path.append('tools/')

import numpy as np
import pandas as pd
from scipy.stats import norm

# Graphing Libraries
import matplotlib.pyplot as pyplt
import seaborn as sns
sns.set_style("white")  


from IPython.display import display
pyplt.rcParams['figure.figsize'] = (4, 3)


    

In [31]:

    

# Simple line graph of the density function of a standard normal.

X = np.linspace(-6, 6, 500)
mean = 0
variance = 1
density = norm.pdf(X, mean, variance)

plot(X, density)
pyplt.title("Gaussian pdf")

mean = 0
variance = 0.5
density = norm.pdf(X, mean, variance)
plot(X, density);


    

In [40]:

    

e = np.exp(1)
x = np.linspace(-10, 10, 1000)
y = e**x / (e**x + 1)
pyplt.plot(x, y)
pyplt.title('sigmoid function');


    

In [89]:

    

from scipy.special import logit
from sklearn.linear_model import LogisticRegression

data = pd.read_csv('data/sat2014.csv')
data.head()


    

    
Out[89]:






  

    

      

      
State
      
Participation Rate
      
Critical Reading
      
Math
      
Writing
      
Combined
    
  
  

    

      
0
      
North Dakota
      
2.3
      
612
      
620
      
584
      
1816
    
    

      
1
      
Illinois
      
4.6
      
599
      
616
      
587
      
1802
    
    

      
2
      
Iowa
      
3.1
      
605
      
611
      
578
      
1794
    
    

      
3
      
South Dakota
      
2.9
      
604
      
609
      
579
      
1792
    
    

      
4
      
Minnesota
      
5.9
      
598
      
610
      
578
      
1786
    
  

In [127]:

    

X, y = data.ix[:, 2:5], data.ix[:, 5]


logistic = LogisticRegression()
model = logistic.fit(X, y)

# Solve for the decision boundary
a = model.coef_; b = model.intercept_;


    

In [ ]:

    




    

In [ ]: