In [41]:

    

%matplotlib inline


    

In [42]:

    

import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import matplotlib.pyplot as plt
seed = 12345
np.random.seed(seed)

from keras.models import Sequential
from keras.layers import Dense
from keras.wrappers.scikit_learn import KerasRegressor
from sklearn.model_selection import cross_val_score
from sklearn.model_selection import KFold
from sklearn.preprocessing import StandardScaler
from sklearn.pipeline import Pipeline


    

In [43]:

    

from keras.datasets import boston_housing
(train_data, train_targets), (test_data, test_targets) = boston_housing.load_data()


    

In [44]:

    

train_data.shape


    

    
Out[44]:





(404, 13)

In [45]:

    

test_data.shape


    

    
Out[45]:





(102, 13)

In [46]:

    

train_targets.shape


    

    
Out[46]:





(404,)

In [47]:

    

test_targets.shape


    

    
Out[47]:





(102,)

In [75]:

    

def baseline_model(n_features=13, n=15, optimizer='adam'):
    # create model
    model = Sequential()
    model.add(Dense(n, input_dim=n_features, kernel_initializer='normal', activation='relu'))
    model.add(Dense(1, kernel_initializer='normal'))
    # Compile model
    model.compile(loss='mean_squared_error', optimizer=optimizer)
    return model


    

In [76]:

    

model = KerasRegressor(build_fn=baseline_model, epochs=100, batch_size=5,verbose=0)


    

In [80]:

    

pipeline = Pipeline([
    ('standard', StandardScaler()),
    ('model', model)
])
kfold = KFold(n_splits=10, random_state=seed)
results = cross_val_score(pipeline, train_data, train_targets, cv=kfold)


    

In [81]:

    

# Cross val score returns -ive mses
results = pd.Series(results).abs()


    

In [82]:

    

results.plot(kind='bar', title='MSE per fold')


    

    
Out[82]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fd499fee668>






    

In [ ]: