Scikit Learn

Scikit Learn is built on Numpy, SciPy and matplotlib for machine learning. This library contains a lot of effecient tools for machine learning and statistical modeling including classification, regression, clustering.

Four step to fit model in Scikit Learn

Prepare the data
Initialize the model
Fit the data
Evaluate the model

Linear Regression Model



In [ ]:

    
from sklearn import linear_model
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline



In [ ]:

    
x_train = np.random.normal(10, 20, 100)
noise = np.random.normal(1, 10, 100)
y_train = x_train * 2.0 + 1.0 + noise

plt.scatter(x_train, y_train)



In [ ]:

    
x_train = x_train.reshape((100, 1))



In [ ]:

    
x_train.shape



In [ ]:

    
linear = linear_model.LinearRegression()  # Initialize model
linear.fit(x_train, y_train)  # fit data
print('Score: \n', linear.score(x_train, y_train)) # Evaluate the model
print('Coefficient: \n', linear.coef_)
print('Intercept: \n', linear.intercept_)
#Predict Output
predicted= linear.predict(x_train)



In [ ]:

    
plt.scatter(x_train, y_train)
x_space = np.linspace(*plt.xlim())
plt.plot(x_space, linear.coef_*x_space + linear.intercept_)

Cross validation



In [ ]:

    
from sklearn.cross_validation import train_test_split
X_train, X_test, y_train, y_test = train_test_split(x_train, y_train, random_state=1)



In [ ]:

    
# default split is 75% for training and 25% for testing
print(X_train.shape)
print(y_train.shape)
print(X_test.shape)
print(y_test.shape)



In [ ]:

    
linear = linear_model.LinearRegression()
linear.fit(X_train, y_train)
print('Score with train data: \n', linear.score(X_train, y_train))
print('Coefficient: \n', linear.coef_)
print('Intercept: \n', linear.intercept_)
#Predict Output
print('Score with test data: \n', linear.score(X_test, y_test))



In [ ]:

    
from IPython.core.display import HTML
HTML("<iframe src=http://scikit-learn.org/stable/  width=800 height=350></iframe>")



In [ ]: