def test_mse_tree():
X, y = boston_housing_data()
X_train, X_test, y_train, y_test = train_test_split(X,
y,
test_size=0.3,
random_state=123,
shuffle=True)
tree = DecisionTreeRegressor(random_state=123)
avg_expected_loss, avg_bias, avg_var = bias_variance_decomp(
tree, X_train, y_train, X_test, y_test, loss='mse', random_seed=123)
assert round(avg_expected_loss, 3) == 31.536
assert round(avg_bias, 3) == 14.096
assert round(avg_var, 3) == 17.440
def test_mse_tree():
X, y = boston_housing_data()
X_train, X_test, y_train, y_test = train_test_split(X, y,
test_size=0.3,
random_state=123,
shuffle=True)
tree = DecisionTreeRegressor(random_state=123)
avg_expected_loss, avg_bias, avg_var = bias_variance_decomp(
tree, X_train, y_train, X_test, y_test,
loss='mse',
random_seed=123)
assert round(avg_expected_loss, 3) == 31.917
assert round(avg_bias, 3) == 13.814
assert round(avg_var, 3) == 18.102
def test_mse_bagging():
X, y = boston_housing_data()
X_train, X_test, y_train, y_test = train_test_split(X,
y,
test_size=0.3,
random_state=123,
shuffle=True)
tree = DecisionTreeRegressor(random_state=123)
bag = BaggingRegressor(base_estimator=tree,
n_estimators=10,
random_state=123)
avg_expected_loss, avg_bias, avg_var = bias_variance_decomp(
bag, X_train, y_train, X_test, y_test, loss='mse', random_seed=123)
assert round(avg_expected_loss, 2) == 20.24, avg_expected_loss
assert round(avg_bias, 2) == 15.63, avg_bias
assert round(avg_var, 2) == 4.61, avg_var
def test_keras():
import tensorflow as tf
X, y = boston_housing_data()
X_train, X_test, y_train, y_test = train_test_split(X,
y,
test_size=0.3,
random_state=123,
shuffle=True)
model = tf.keras.Sequential([
tf.keras.layers.Dense(32, activation=tf.nn.relu),
tf.keras.layers.Dense(1)
])
optimizer = tf.keras.optimizers.Adam()
model.compile(loss='mean_squared_error', optimizer=optimizer)
model.fit(X_train, y_train, epochs=10)
model.predict(X_test)
def test_mse_bagging():
X, y = boston_housing_data()
X_train, X_test, y_train, y_test = train_test_split(X, y,
test_size=0.3,
random_state=123,
shuffle=True)
tree = DecisionTreeRegressor(random_state=123)
bag = BaggingRegressor(base_estimator=tree,
n_estimators=100,
random_state=123)
avg_expected_loss, avg_bias, avg_var = bias_variance_decomp(
bag, X_train, y_train, X_test, y_test,
loss='mse',
random_seed=123)
assert round(avg_expected_loss, 3) == 18.593
assert round(avg_bias, 3) == 15.354
assert round(avg_var, 3) == 3.239
def test_regressor():
X, y = boston_housing_data()
reg1 = Lasso(random_state=1)
reg2 = Ridge(random_state=1)
X_train, X_test, y_train, y_test = \
train_test_split(X, y, test_size=0.25,
random_state=123)
score1 = reg1.fit(X_train, y_train).score(X_test, y_test)
score2 = reg2.fit(X_train, y_train).score(X_test, y_test)
assert round(score1, 2) == 0.66, score1
assert round(score2, 2) == 0.68, score2
t, p = paired_ttest_kfold_cv(estimator1=reg1,
estimator2=reg2,
X=X, y=y,
random_seed=1)
assert round(t, 3) == -0.549, t
assert round(p, 3) == 0.596, p
def test_regressor():
X, y = boston_housing_data()
reg1 = Lasso(random_state=1)
reg2 = Ridge(random_state=1)
X_train, X_test, y_train, y_test = \
train_test_split(X, y, test_size=0.25,
random_state=123)
score1 = reg1.fit(X_train, y_train).score(X_test, y_test)
score2 = reg2.fit(X_train, y_train).score(X_test, y_test)
assert round(score1, 2) == 0.66, score1
assert round(score2, 2) == 0.68, score2
t, p = paired_ttest_5x2cv(estimator1=reg1,
estimator2=reg2,
X=X, y=y,
random_seed=1)
assert round(t, 3) == -0.599, t
assert round(p, 3) == 0.575, p
def test_regressor():
X, y = boston_housing_data()
reg1 = Lasso(random_state=1)
reg2 = Ridge(random_state=1)
X_train, X_test, y_train, y_test = \
train_test_split(X, y, test_size=0.25,
random_state=123)
score1 = reg1.fit(X_train, y_train).score(X_test, y_test)
score2 = reg2.fit(X_train, y_train).score(X_test, y_test)
assert round(score1, 2) == 0.66, score1
assert round(score2, 2) == 0.68, score2
f, p = combined_ftest_5x2cv(estimator1=reg1,
estimator2=reg2,
X=X, y=y,
random_seed=1)
assert round(f, 3) == 3.211, f
assert round(p, 3) == 0.105, p
def test_regressor():
X, y = boston_housing_data()
reg1 = Lasso(random_state=1)
reg2 = Ridge(random_state=1)
X_train, X_test, y_train, y_test = \
train_test_split(X, y, test_size=0.25,
random_state=123)
score1 = reg1.fit(X_train, y_train).score(X_test, y_test)
score2 = reg2.fit(X_train, y_train).score(X_test, y_test)
assert round(score1, 2) == 0.66, score1
assert round(score2, 2) == 0.68, score2
f, p = combined_ftest_5x2cv(estimator1=reg1,
estimator2=reg2,
X=X,
y=y,
random_seed=1)
assert round(f, 3) == 3.211, f
assert round(p, 3) == 0.105, p
from mlxtend.regressor import StackingCVRegressor
from mlxtend.data import boston_housing_data
from sklearn.linear_model import LinearRegression, Ridge, Lasso
from sklearn.svm import SVR
from sklearn.model_selection import train_test_split, GridSearchCV, cross_val_score
from sklearn.metrics import mean_squared_error
import numpy as np
import matplotlib.pyplot as plt
x, y = boston_housing_data()
x = x[:100]
y = y[:100]
# 划分数据集
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2)
# 初始化基模型
lr = LinearRegression()
svr_lin = SVR(kernel='linear', gamma='auto')
ridge = Ridge(random_state=2019, )
lasso = Lasso()
models = [lr, svr_lin, ridge, lasso]
print("base model")
for model in models:
score = cross_val_score(model, x_train, y_train, cv=5)
print(score.mean(), "+/-", score.std())
sclf = StackingCVRegressor(regressors=models, meta_regressor=lasso)
# 训练回归器
print("stacking model")
score = cross_val_score(sclf, x_train, y_train, cv=5)
print(score.mean(), "+/-", score.std())
def test_import_boston_housing():
X, y = boston_housing_data()
assert (X.shape[0] == 506)
assert (X.shape[1] == 13)
assert (y.shape[0] == 506)
def test_import_boston_housing():
X, y = boston_housing_data()
assert(X.shape[0] == 506)
assert(X.shape[1] == 13)
assert(y.shape[0] == 506)
# Sebastian Raschka 2014-2018
# mlxtend Machine Learning Library Extensions
# Author: Sebastian Raschka <sebastianraschka.com>
#
# License: BSD 3 clause
from mlxtend.regressor import LinearRegression
from mlxtend.data import boston_housing_data
import numpy as np
from numpy.testing import assert_almost_equal
from sklearn.base import clone
X, y = boston_housing_data()
X_rm = X[:, 5][:, np.newaxis]
X_rm_lstat = X[:, [5, -1]]
# standardized variables
X_rm_std = (X_rm - X_rm.mean(axis=0)) / X_rm.std(axis=0)
X_rm_lstat_std = ((X_rm_lstat - X_rm_lstat.mean(axis=0)) /
X_rm_lstat.std(axis=0))
y_std = (y - y.mean()) / y.std()
def test_univariate_normal_equation():
w_exp = np.array([[9.1]])
b_exp = np.array([-34.7])
ne_lr = LinearRegression(minibatches=None)
ne_lr.fit(X_rm, y)
assert_almost_equal(ne_lr.w_, w_exp, decimal=1)
assert_almost_equal(ne_lr.b_, b_exp, decimal=1)
