Title: Make Simulated Data For Regression
Slug: make_simulated_data_for_regression
Summary: Make a simulated dataset for regression using scikit-learn.
Date: 2017-01-16 12:00
Category: Machine Learning
Tags: Basics
Authors: Chris Albon

Preliminaries



In [1]:

    
import pandas as pd
from sklearn.datasets import make_regression

Create Simulated Data



In [2]:

    
# Generate fetures, outputs, and true coefficient of 100 samples,
features, output, coef = make_regression(n_samples = 100,
                                         # three features
                                         n_features = 3,
                                         # where only two features are useful,
                                         n_informative = 2,
                                         # a single target value per observation
                                         n_targets = 1,
                                         # 0.0 standard deviation of the guassian noise
                                         noise = 0.0,
                                         # show the true coefficient used to generated the data
                                         coef = True)

View Simulated Data



In [3]:

    
# View the features of the first five rows
pd.DataFrame(features, columns=['Store 1', 'Store 2', 'Store 3']).head()



In [4]:

    
# View the output of the first five rows
pd.DataFrame(output, columns=['Sales']).head()









    Out[4]:






  
    
      
      Sales
    
  
  
    
      0
      -149.387162
    
    
      1
      -4.164344
    
    
      2
      52.166904
    
    
      3
      -56.996180
    
    
      4
      27.246575



In [5]:

    
# View the actual, true coefficients used to generate the data
pd.DataFrame(coef, columns=['True Coefficient Values'])









    Out[5]:






  
    
      
      True Coefficient Values
    
  
  
    
      0
      0.000000
    
    
      1
      80.654346
    
    
      2
      57.993548

	Store 1	Store 2	Store 3
0	-0.166697	-0.177142	-2.329568
1	-0.093566	-0.544292	0.685165
2	0.625958	-0.193049	1.168012
3	-0.843925	-0.567444	-0.193631
4	-1.079227	-0.819236	1.609171