stacking_example

Complexity

Total Complexity

6

Size/Duplication

Total Lines	88
Duplicated Lines	0 %

Importance

Changes

0

3 Functions

Rating	Name	Duplication	Size	Complexity
A	warn()	0	2	1
A	get_combinations()	0	8	4
A	stacking()	0	10	1

# disables sklearn warnings
def warn(*args, **kwargs):
    pass


import warnings

warnings.warn = warn

import itertools

from sklearn.datasets import load_breast_cancer
from sklearn.model_selection import cross_val_score
from mlxtend.classifier import StackingClassifier

from sklearn.ensemble import (
    GradientBoostingClassifier,
    RandomForestClassifier,
    ExtraTreesClassifier,
)

from sklearn.neighbors import KNeighborsClassifier
from sklearn.neural_network import MLPClassifier
from sklearn.gaussian_process import GaussianProcessClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.naive_bayes import GaussianNB

from sklearn.linear_model import LogisticRegression
from sklearn.linear_model.ridge import RidgeClassifier

from hyperactive import Hyperactive

data = load_breast_cancer()
X, y = data.data, data.target


gbc = GradientBoostingClassifier()
rfc = RandomForestClassifier()
etc = ExtraTreesClassifier()

mlp = MLPClassifier()
gnb = GaussianNB()
gpc = GaussianProcessClassifier()
dtc = DecisionTreeClassifier()
knn = KNeighborsClassifier()

lr = LogisticRegression()
rc = RidgeClassifier()


def stacking(para, X, y):
    stack_lvl_0 = StackingClassifier(
        classifiers=para["lvl_0"], meta_classifier=para["top"]
    )
    stack_lvl_1 = StackingClassifier(
        classifiers=para["lvl_1"], meta_classifier=stack_lvl_0
    )
    scores = cross_val_score(stack_lvl_1, X, y, cv=3)

    return scores.mean()


def get_combinations(models):
    comb = []
    for i in range(0, len(models) + 1):
        for subset in itertools.permutations(models, i):
            if len(subset) == 0:
                continue
            comb.append(list(subset))
    return comb


top = [lr, dtc, gnb, rc]
models_0 = [gpc, dtc, mlp, gnb, knn]
models_1 = [gbc, rfc, etc]

stack_lvl_0_clfs = get_combinations(models_0)
stack_lvl_1_clfs = get_combinations(models_1)


search_config = {
    stacking: {"lvl_1": stack_lvl_1_clfs, "lvl_0": stack_lvl_0_clfs, "top": top}
}


opt = Hyperactive(search_config, n_jobs=2, n_iter=150)
opt.search(X, y)