Hyperparameter tuning of Machine learning models using manual tuning

Manual_tuning.py

#import libraries
from sklearn import datasets
from catboost import CatBoostClassifier
from sklearn.model_selection import cross_val_score

# load data
cancer = datasets.load_breast_cancer()

# target
y = cancer.target

# features
X = cancer.data

#Instantiate CatBoostClassifier, using a maximum depth of 3
cbc = CatBoostClassifier(max_depth=3)

# 5 folds, scored on accuracy
cvs = cross_val_score(cbc, X, y, cv=5, scoring='accuracy')

#Mean value of cross validation score
print (f 'The mean value of cross val score is {cvs.mean()}') #mean = 0.96

print("======="*5)

#Instantiate CatBoostClassifier, using a maximum depth of 5
cbc1 = CatBoostClassifier(max_depth=5)

# 5 folds, scored on accuracy
cvs1 = cross_val_score(cbc1, X, y, cv=5, scoring='accuracy')

#Mean value of cross validation score
print (f 'The mean value of cross val score is {cvs1.mean()}') #mean = 0.97