In [5]:

    

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
plt.style.use("ggplot")


    

In [3]:

    

%matplotlib` inline


    

In [41]:

    

data = pd.read_csv("http://bit.do/df-housing")


    

In [44]:

    

df = data[['SalePrice', 'OverallQual', 'GrLivArea', 'GarageCars', 'TotalBsmtSF', 'FullBath', 'YearBuilt', 'FireplaceQu', 'LotFrontage']]


    

In [49]:

    

df.head()


    

    
Out[49]:







  

    

      

      
SalePrice
      
OverallQual
      
GrLivArea
      
GarageCars
      
TotalBsmtSF
      
FullBath
      
YearBuilt
      
FireplaceQu
      
LotFrontage
    
  
  

    

      
0
      
208500
      
7
      
1710
      
2
      
856
      
2
      
2003
      
NaN
      
65.0
    
    

      
1
      
181500
      
6
      
1262
      
2
      
1262
      
2
      
1976
      
TA
      
80.0
    
    

      
2
      
223500
      
7
      
1786
      
2
      
920
      
2
      
2001
      
TA
      
68.0
    
    

      
3
      
140000
      
7
      
1717
      
3
      
756
      
1
      
1915
      
Gd
      
60.0
    
    

      
4
      
250000
      
8
      
2198
      
3
      
1145
      
2
      
2000
      
TA
      
84.0
    
  

In [56]:

    

X_raw = df.drop('SalePrice', axis=1)
y_raw = df.SalePrice


    

In [59]:

    

from sklearn.pipeline import Pipeline
from sklearn.compose import ColumnTransformer
from sklearn.preprocessing import OneHotEncoder, StandardScaler
from sklearn.impute import SimpleImputer


    

In [73]:

    

cat_cols = ['FireplaceQu']
num_cols = ['GrLivArea', 'GarageCars', 'TotalBsmtSF', 'FullBath', 'YearBuilt', "OverallQual" ,'LotFrontage']


    

In [74]:

    

cat_si_step = ('si', SimpleImputer(strategy='constant', fill_value='MISSING'))
cat_ohe_step = ('ohe', OneHotEncoder(sparse=False, handle_unknown='ignore'))
cat_steps = [cat_si_step, cat_ohe_step]
cat_pipe = Pipeline(cat_steps)
cat_transformers = [('cat', cat_pipe, cat_cols)]


    

In [75]:

    

num_si_step = ('si', SimpleImputer(strategy='median'))
num_ss_step = ('ss', StandardScaler())
num_steps = [num_si_step, num_ss_step]
num_pipe = Pipeline(num_steps)
num_transformers = [('num', num_pipe, num_cols)]


    

In [76]:

    

transformers = [('cat', cat_pipe, cat_cols),
                ('num', num_pipe, num_cols)]
ct = ColumnTransformer(transformers=transformers)
X_encoded = ct.fit_transform(X)
X_encoded.shape


    

    
Out[76]:





(1460, 13)

In [78]:

    

from sklearn.linear_model import Ridge


    

In [79]:

    

ml_pipe = Pipeline([('transform', ct), ('ridge', Ridge())])
ml_pipe.fit(X, y)


    

    
Out[79]:





Pipeline(memory=None,
     steps=[('transform', ColumnTransformer(n_jobs=None, remainder='drop', sparse_threshold=0.3,
         transformer_weights=None,
         transformers=[('cat', Pipeline(memory=None,
     steps=[('si', SimpleImputer(copy=True, fill_value='MISSING', missing_values=nan,
       strategy='constant', verbos...it_intercept=True, max_iter=None,
   normalize=False, random_state=None, solver='auto', tol=0.001))])

In [81]:

    

ml_pipe.score(X, y)


    

    
Out[81]:





0.7772097032829757

In [83]:

    

from sklearn.model_selection import KFold, cross_val_score
kf = KFold(n_splits=5, shuffle=True, random_state=123)
cross_val_score(ml_pipe, X, y, cv=kf).mean()


    

    
Out[83]:





0.7550365007527828

In [85]:

    

from sklearn.model_selection import GridSearchCV
param_grid = {
    'transform__num__si__strategy': ['mean', 'median'],
    'ridge__alpha': [.001, 0.1, 1.0, 5, 10, 50, 100, 1000],
    }
gs = GridSearchCV(ml_pipe, param_grid, cv=kf)
gs.fit(X, y)
gs.best_params_


    

    
Out[85]:





{'ridge__alpha': 50, 'transform__num__si__strategy': 'median'}

In [ ]: