Stacking in Machine Learning

Stacking:
Stacking is a way of ensembling classification or regression models it consists of two-layer estimators. The first layer consists of all the baseline models that are used to predict the outputs on the test datasets. The second layer consists of Meta-Classifier or Regressor which takes all the predictions of baseline models as an input and generate new predictions.
Stacking Architecture:

Stacking Architecture

mlxtend:
Mlxtend (machine learning extensions) is a Python library of useful tools for day-to-day data science tasks. It consists of lots of tools that are useful for data science and machine learning tasks for example: 

1. Feature Selection
2. Feature Extraction
3. Visualization
4. Ensembling

and many more.
This article explains how to implement Stacking Classifier on the classification dataset.
Why Stacking? 
Most of the Machine-Learning and Data science competitions are won by using Stacked models. They can improve the existing accuracy that is shown by individual models. We can get most of the Stacked models by choosing diverse algorithms in the first layer of architecture as different algorithms capture different trends in training data by combining both of the models can give better and accurate results.
Installation of libraries on the system:

pip install mlxtend 
pip install pandas 
pip install -U scikit-learn

Code: Import Required Libraries:

python3

import pandas as pd
import matplotlib.pyplot as plt
from mlxtend.plotting import plot_confusion_matrix
from mlxtend.classifier import StackingClassifier
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import GaussianNB
from sklearn.metrics import confusion_matrix
from sklearn.metrics import accuracy_score

Code: Loading the dataset

python3

df = pd.read_csv('heart.csv')    # loading the dataset
df.head()                        # viewing top 5 rows of dataset

Output:

Code: 

python3

# Creating X and y for training
X = df.drop('target', axis = 1)
y = df['target']

Code: Splitting Data into Train and Test

python3

# 20 % training dataset is considered for testing
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 42)

Code: Standardizing Data

python3

# initializing sc object
sc = StandardScaler()
# variables that needed to be transformed
var_transform = ['thalach', 'age', 'trestbps', 'oldpeak', 'chol']
X_train[var_transform] = sc.fit_transform(X_train[var_transform])   # standardizing training data
X_test[var_transform] = sc.transform(X_test[var_transform])            # standardizing test data
print(X_train.head())

Output:

Code: Building First Layer Estimators

python3

KNC = KNeighborsClassifier()   # initialising KNeighbors Classifier
NB = GaussianNB()              # initialising Naive Bayes

Let’s Train and evaluate with our first layer estimators to observe the difference in the performance of the stacked model and general model
Code: Training KNeighborsClassifier

python3

model_kNeighborsClassifier = KNC.fit(X_train, y_train)   # fitting Training Set
pred_knc = model_kNeighborsClassifier.predict(X_test)   # Predicting on test dataset

Code: Evaluation of KNeighborsClassifier

python3

acc_knc = accuracy_score(y_test, pred_knc)  # evaluating accuracy score
print('accuracy score of KNeighbors Classifier is:', acc_knc * 100)

Output:

Code: Training Naive Bayes Classifier

python3

model_NaiveBayes = NB.fit(X_train, y_train)
pred_nb = model_NaiveBayes.predict(X_test)

Code: Evaluation of Naive Bayes Classifier

python3

acc_nb = accuracy_score(y_test, pred_nb)
print('Accuracy of Naive Bayes Classifier:', acc_nb * 100)

Output:

Code: Implementing Stacking Classifier

python3

lr = LogisticRegression()  # defining meta-classifier
clf_stack = StackingClassifier(classifiers =[KNC, NB], meta_classifier = lr, use_probas = True, use_features_in_secondary = True)

• use_probas=True indicates the Stacking Classifier uses the prediction probabilities as an input instead of using predictions classes.
• use_features_in_secondary=True indicates Stacking Classifier not only take predictions as an input but also uses features in the dataset to predict on new data.

Code: Training Stacking Classifier

python3

model_stack = clf_stack.fit(X_train, y_train)   # training of stacked model
pred_stack = model_stack.predict(X_test)       # predictions on test data using stacked model

Code: Evaluating Stacking Classifier

python3

acc_stack = accuracy_score(y_test, pred_stack)  # evaluating accuracy
print('accuracy score of Stacked model:', acc_stack * 100)

Output:

Our both individual models scores an accuracy of nearly 80% and our Stacked model got an accuracy of nearly 84%.By Combining two individual models we got a significant performance improvement.
Code: 

python3

model_stack = clf_stack.fit(X_train, y_train)   # training of stacked model
pred_stack = model_stack.predict(X_test)       # predictions on test data using stacked model

Code: Evaluating Stacking Classifier

python3

acc_stack = accuracy_score(y_test, pred_stack)  # evaluating accuracy
print('accuracy score of Stacked model:', acc_stack * 100)

Output:

Our both individual models scores an accuracy of nearly 80% and our Stacked model got an accuracy of nearly 84%. By Combining two individual models we got a significant performance improvement.