67 lines
2.6 KiB
Python
67 lines
2.6 KiB
Python
|
from sklearn.feature_selection import f_regression
|
||
|
|
from sklearn.linear_model import LinearRegression, Ridge, Lasso
|
||
|
|
from sklearn.preprocessing import MinMaxScaler
|
||
|
|
import numpy as np
|
||
|
|
|
||
|
|
def rank_to_dict(ranks, names):
|
||
|
|
ranks = np.abs(ranks)
|
||
|
|
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))
|
||
|
|
|
||
|
|
def mean_ranks(ranks):
|
||
|
|
mean = {} # словарь для хранения средних рангов признаков
|
||
|
|
for key, value in ranks.items():
|
||
|
|
for item in value.items():
|
||
|
|
if item[0] not in mean:
|
||
|
|
mean[item[0]] = 0
|
||
|
|
mean[item[0]] += item[1]
|
||
|
|
|
||
|
|
for key, value in mean.items():
|
||
|
|
res = value / len(ranks)
|
||
|
|
mean[key] = round(res, 2)
|
||
|
|
|
||
|
|
sorted_mean = sorted(mean.items(), key=lambda x: x[1], reverse=True)
|
||
|
|
return sorted_mean
|
||
|
|
|
||
|
|
|
||
|
|
# генерируем исходные данные: 750 строк-наблюдений и 14 столбцов-признаков
|
||
|
|
np.random.seed(0)
|
||
|
|
size = 750
|
||
|
|
X = np.random.uniform(0, 1, (size, 14))
|
||
|
|
# задаем функцию-выход: регрессионную проблему Фридмана
|
||
|
|
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))
|
||
|
|
# добавляем зависимость признаков
|
||
|
|
X[:,10:] = X[:,:4] + np.random.normal(0, .025, (size,4))
|
||
|
|
|
||
|
|
# создаем модели
|
||
|
|
linear_regression = LinearRegression()
|
||
|
|
ridge = Ridge()
|
||
|
|
lasso = Lasso(alpha=.05)
|
||
|
|
f, pval = f_regression(X, Y, center=True)
|
||
|
|
models = [("Линейная регрессия", linear_regression),
|
||
|
|
("Гребневая регрессия", ridge),
|
||
|
|
("Лассо", lasso),
|
||
|
|
("Линейная корреляция", f)]
|
||
|
|
|
||
|
|
names = ["x%s" % i for i in range(1,15)] # список содержащий, имена признаков
|
||
|
|
ranks = {} # словарь для хранения рангов признаков
|
||
|
|
|
||
|
|
for (name, model) in models:
|
||
|
|
if name != "Линейная корреляция":
|
||
|
|
model.fit(X, Y) # тренируем модель
|
||
|
|
ranks[name] = rank_to_dict(model.coef_, names) # преобразуем признаки в словарь
|
||
|
|
else:
|
||
|
|
ranks[name] = rank_to_dict(model, names)
|
||
|
|
|
||
|
|
print("-----ОЦЕНКИ ПРИЗНАКОВ-----")
|
||
|
|
for key, value in ranks.items():
|
||
|
|
print(key, ":")
|
||
|
|
for item in value.items():
|
||
|
|
print(item[0], "-", item[1])
|
||
|
|
print()
|
||
|
|
|
||
|
|
print("---СРЕДНИЕ ОЦЕНКИ ПРИЗНАКОВ---")
|
||
|
|
for r in mean_ranks(ranks):
|
||
|
|
print(r[0], "-", r[1])
|