ML Regression

概述：ML Regression

[TOC]

1、线性回归

# 加载线性模型算法库
from sklearn import linear_model 
# 创建线性回归模型的对象 
regr = linear_model.LinearRegression() 
# 利用训练集训练线性模型 
regr.fit(X_train, y_train) 
# 使用测试集做预测 
y_pred = regr.predict(X_test)

2、岭回归

对系数进行惩罚(L2范式)来解决普通最小二乘法

#加载线性模型算法库 
from sklearn.linear_model import Ridge 
# 创建岭回归模型的对象 reg = Ridge(alpha=.5) 
# 利用训练集训练岭回归模型 
reg.fit([[0, 0], [0, 0], [1, 1]], [0, .1, 1]) 
#输出各个系数 
reg.coef_ reg.intercept_

3、Lasso回归

最小二乘法的基础上加入L1范式作为惩罚项

#加载Lasso模型算法库 
from sklearn.linear_model import Lasso 
# 创建Lasso回归模型的对象 
reg = Lasso(alpha=0.1) 
# 利用训练集训练Lasso回归模型 
reg.fit([[0, 0], [1, 1]], [0, 1]) 
""" Lasso(alpha=0.1, copy_X=True, fit_intercept=True, max_iter=1000, normalize=False, positive=False, precompute=False, random_state=None, selection='cyclic', tol=0.0001, warm_start=False) """ 
# 使用测试集做预测 
reg.predict([[1, 1]])

4、Elastic Net回归

利用L1范式和L2范式共同作为惩罚项

#加载ElasticNet模型算法库 
from sklearn.linear_model import ElasticNet 
#加载数据集 
from sklearn.datasets import make_regression 
X, y = make_regression(n_features=2, random_state=0) 
#创建ElasticNet回归模型的对象 
regr = ElasticNet(random_state=0) 
# 利用训练集训练ElasticNet回归模型 
regr.fit(X, y) 
print(regr.coef_) 
print(regr.intercept_) 
print(regr.predict([[0, 0]]))

5、贝叶斯岭回归

贝叶斯岭回归模型和岭回归类似。贝叶斯岭回归通过最大化边际对数似然来估计参数。

from sklearn.linear_model import BayesianRidge 
X = [[0., 0.], [1., 1.], [2., 2.], [3., 3.]] 
Y = [0., 1., 2., 3.] 
reg = BayesianRidge() reg.fit(X, Y)

6、SGD回归

上述的线性模型通过最小二乘法来优化损失函数，SGD回归也是一种线性回归，不同的是，它通过随机梯度下降最小化正则化经验损失。

import numpy as np 
from sklearn import linear_model 
n_samples, n_features = 10, 5 
np.random.seed(0) 
y = np.random.randn(n_samples) 
X = np.random.randn(n_samples, n_features) 
clf = linear_model.SGDRegressor(max_iter=1000, tol=1e-3) clf.fit(X, y)

7、SVR

#加载SVR模型算法库 
from sklearn.svm import SVR 
#训练集 
X = [[0, 0], [2, 2]] 
y = [0.5, 2.5] 
#创建SVR回归模型的对象 
clf = SVR() 
# 利用训练集训练SVR回归模型 
clf.fit(X, y) 
""" SVR(C=1.0, cache_size=200, coef0=0.0, degree=3, epsilon=0.1, gamma='auto_deprecated', kernel='rbf', max_iter=-1, shrinking=True, tol=0.001, verbose=False) """ 
clf.predict([[1, 1]])

8、KNN回归

X = [[0], [1], [2], [3]] 
y = [0, 0, 1, 1] 
from sklearn.neighbors import KNeighborsRegressor 
neigh = KNeighborsRegressor(n_neighbors=2) 
neigh.fit(X, y) print(neigh.predict([[1.5]]))

9、决策树回归

from sklearn.tree import DecisionTreeRegressor 
X = [[0, 0], [2, 2]] 
y = [0.5, 2.5] 
clf = DecisionTreeRegressor() 
clf = clf.fit(X, y) clf.predict([[1, 1]])

10、神经网络

1
2
3

from sklearn.neural_network import MLPRegressor mlp=MLPRegressor() 
mlp.fit(X_train,y_train) 
y_pred = mlp.predict(X_test)

11.RandomForest回归

from sklearn.ensemble import RandomForestRegressor 
from sklearn.datasets import make_regression 
X, y = make_regression(n_features=4, n_informative=2, random_state=0, shuffle=False) 
regr = RandomForestRegressor(max_depth=2, random_state=0, n_estimators=100) 
regr.fit(X, y) 
print(regr.feature_importances_) 
print(regr.predict([[0, 0, 0, 0]]))

12、XGBoost回归

1
2
3

import xgboost as xgb 
xgb_model = xgb.XGBRegressor(max_depth = 3, learning_rate = 0.1, n_estimators = 100, objective = 'reg:linear', n_jobs = -1) 
xgb_model.fit(X_train, y_train, eval_set=[(X_train, y_train)], eval_metric='logloss', verbose=100) y_pred = xgb_model.predict(X_test) print(mean_squared_error(y_test, y_pred))

13、LightGBM回归

LightGBM作为另一个使用基于树的学习算法的梯度增强框架。相比于XGBoost，LightGBM有如下优点，训练速度更快，效率更高效；低内存的使用量，优势：更快的训练效率，低内存使用

更好的准确率；支持并行学习；可处理大规模数据

import lightgbm as lgb 
gbm = lgb.LGBMRegressor(num_leaves=31, learning_rate=0.05, n_estimators=20) 
gbm.fit(X_train, y_train, eval_set=[(X_train, y_train)], eval_metric='logloss', verbose=100) 
y_pred = gbm.predict(X_test) print(mean_squared_error(y_test, y_pred))