81 lines
3.0 KiB
Python
81 lines
3.0 KiB
Python
from sklearn.linear_model import LinearRegression, Lasso
|
|
from sklearn.feature_selection import RFE
|
|
from sklearn.preprocessing import MinMaxScaler
|
|
import numpy as np
|
|
|
|
|
|
# Генерация синтетических данных
|
|
def create_data():
|
|
np.random.seed(0)
|
|
size = 750
|
|
X = np.random.uniform(0, 1, (size, 14))
|
|
Y = (10 * np.sin(np.pi * X[:, 0] * X[:, 1]) + 20 * (X[:, 2] - .5) ** 2 +
|
|
10 * X[:, 3] + 5 * X[:, 4] ** 5 + np.random.normal(0, 1))
|
|
X[:, 10:] = X[:, :4] + np.random.normal(0, .025, (size, 4))
|
|
return X, Y
|
|
|
|
|
|
# Нормализация значений рангов
|
|
def rank_to_dict(ranks):
|
|
ranks = np.abs(ranks)
|
|
names = ["x%s" % i for i in range(1, 15)]
|
|
minmax = MinMaxScaler()
|
|
ranks = minmax.fit_transform(np.array(ranks).reshape(14, 1)).ravel()
|
|
ranks = map(lambda x: round(x, 2), ranks)
|
|
return dict(zip(names, ranks))
|
|
|
|
|
|
# Вывод отсортированных признаков по важности в табличном виде
|
|
def print_sorted_features_by_models(ranks_by_model: {}):
|
|
sorted_ranks = sorted(ranks_by_model.items(), key=lambda item: sum(item[1].values()), reverse=True)
|
|
|
|
print("{:<40}".format(""), end="")
|
|
for i in range(1, 15):
|
|
print("{:<10}".format(i), end="")
|
|
print()
|
|
|
|
for model, rank_dict in sorted_ranks:
|
|
sorted_features = sorted(rank_dict.items(), key=lambda item: item[1], reverse=True)
|
|
print("{:<40}".format(model), end="")
|
|
for feature, rank in sorted_features:
|
|
print("{:<10}".format(f"{feature}: {rank}"), end="")
|
|
print()
|
|
print()
|
|
|
|
|
|
# Получение средних значений моделей и ТОП 4 самых важных признака
|
|
def average_values(ranks_by_model: {}):
|
|
mean = {}
|
|
for model, rank_dict in ranks_by_model.items():
|
|
for feature, rank in rank_dict.items():
|
|
if feature not in mean:
|
|
mean[feature] = 0
|
|
mean[feature] += rank
|
|
mean = {feature: round(rank / len(ranks_by_model), 2) for feature, rank in mean.items()}
|
|
mean_sorted = sorted(mean.items(), key=lambda item: item[1], reverse=True)
|
|
print("Средние значения")
|
|
print(mean_sorted)
|
|
print("\nТОП 4 самых важных признака по среднему значению: ")
|
|
for feature, rank in mean_sorted[:4]:
|
|
print('Признак - {0}, значение важности - {1}'.format(feature, rank))
|
|
|
|
|
|
X, Y = create_data()
|
|
# ЛИНЕЙНАЯ РЕГРЕССИЯ
|
|
linear_regression = LinearRegression()
|
|
linear_regression.fit(X, Y)
|
|
# ЛАССО
|
|
lasso = Lasso(alpha=.01)
|
|
lasso.fit(X, Y)
|
|
# РЕКУРСИВНОЕ СОКРАЩЕНИЕ ПРИЗНАКОВ
|
|
rfe = RFE(linear_regression)
|
|
rfe.fit(X, Y)
|
|
|
|
ranks_by_model = {
|
|
"Линейная регрессия": rank_to_dict(linear_regression.coef_),
|
|
"Лассо": rank_to_dict(lasso.coef_),
|
|
"РЕКУРСИВНОЕ СОКРАЩЕНИЕ ПРИЗНАКОВ (RFE)": rank_to_dict(rfe.ranking_),
|
|
}
|
|
print_sorted_features_by_models(ranks_by_model)
|
|
average_values(ranks_by_model)
|