What are tune ensemble and blend models functions in PyCaret

Recipe Objective - What are the tune_model, ensemble_model, and blend_model functions in the classification model in PyCaret?

PyCaret provides tune_model, ensemble_model and blend_models functions in the classification module.

For more related projects:-

https://www.dezyre.com/projects/data-science-projects/data-science-projects-in-python
https://www.dezyre.com/projects/data-science-projects/machine-learning-projects-in-python

Tune Model function with Example:

PyCaret provides "pycaret.classification.tune_model()" function. Tune model function tunes the hyperparameters of a given estimator.

from pycaret.datasets import get_data
iris = get_data('iris')
# importing classification module
from pycaret.classification import *
# initialize setup
setup_name = setup(data = iris, target = 'species')
# logistic regression
log_reg = create_model('lr')
tuned_log_reg = tune_model(log_reg)
tuned_log_reg


ensemble_model function with Example:

PyCaret provides "pycaret.classification.ensemble_models()" function. Ensemble model function ensembles a given estimator.

The output of this function is a scoring grid with CV scores by fold.

from pycaret.datasets import get_data
iris = get_data('iris')
# importing classification module
from pycaret.classification import *
# initialize setup
setup_name = setup(data = iris, target = 'species')
# SVM model
svm = create_model('svm')
bagged_svm = ensemble_model(svm, method = 'Bagging')
bagged_svm


blend_models function with Example:

PyCaret provides "pycaret.classification.blend_models()" function. blend_models function trains a Soft Voting / Majority Rule classifier for select models passed in the "estimator_list" parameters.

The output of this function is a scoring grid with CV scores by fold.

from pycaret.datasets import get_data
iris = get_data('iris')
# importing classification module
from pycaret.classification import *
# initialize setup
setup_name = setup(data = iris, target = 'species')
top3 = compare_models(n_select = 3)
blender = blend_models(top3)
blender