目录
6-7 scikit-learn中的随机梯度下降法
封装我们自己的SGD
所有的样本看一边,无法保证有的样本不能使用,将其索引打乱,全
三分之一的够用,两倍的样本量肯定够用了
LinearRegression.py
import numpy as np
from .metrics import r2_score
class LinearRegression:
def __init__(self):
"""初始化Linear Regression模型"""
self.coef_ = None
self.intercept_ = None
self._theta = None
def fit_normal(self, X_train, y_train):
"""根据训练数据集X_train, y_train训练Linear Regression模型"""
assert X_train.shape[0] == y_train.shape[0], \
"the size of X_train must be equal to the size of y_train"
X_b = np.hstack([np.ones((len(X_train), 1)), X_train])
self._theta = np.linalg.inv(X_b.T.dot(X_b)).dot(X_b.T).dot(y_train)
self.intercept_ = self._theta[0]
self.coef_ = self._theta[1:]
return self
def fit_bgd(self, X_train, y_train, eta=0.01, n_iters=1e4):
"""根据训练数据集X_train, y_train, 使用梯度下降法训练Linear Regression模型"""
assert X_train.shape[0] == y_train.shape[0], \
"the size of X_train must be equal to the size of y_train"
def J(theta, X_b, y):
try:
return np.sum((y - X_b.dot(theta)) ** 2) / len(y)
except:
return float('inf')
def dJ(theta, X_b, y):
return X_b.T.dot(X_b.dot(theta) - y) * 2. / len(y)
def gradient_descent(X_b, y, initial_theta, eta, n_iters=1e4, epsilon=1e-8):
theta = initial_theta
cur_iter = 0
while cur_iter < n_iters:
gradient = dJ(theta, X_b, y)
last_theta = theta
theta = theta - eta * gradient
if (abs(J(theta, X_b, y) - J(last_theta, X_b, y)) < epsilon):
break
cur_iter += 1
return theta
X_b = np.hstack([np.ones((len(X_train), 1)), X_train])
initial_theta = np.zeros(X_b.shape[1])
self._theta = gradient_descent(X_b, y_train, initial_theta, eta, n_iters)
self.intercept_ = self._theta[0]
self.coef_ = self._theta[1:]
return self
def fit_sgd(self, X_train, y_train, n_iters=50, t0=5, t1=50):
"""根据训练数据集X_train, y_train, 使用梯度下降法训练Linear Regression模型"""
assert X_train.shape[0] == y_train.shape[0], \
"the size of X_train must be equal to the size of y_train"
assert n_iters >= 1
def dJ_sgd(theta, X_b_i, y_i):
return X_b_i * (X_b_i.dot(theta) - y_i) * 2.
def sgd(X_b, y, initial_theta, n_iters=5, t0=5, t1=50):
def learning_rate(t):
return t0 / (t + t1)
theta = initial_theta
m = len(X_b)
for i_iter in range(n_iters):
indexes = np.random.permutation(m)
X_b_new = X_b[indexes,:]
y_new = y[indexes]
for i in range(m):
gradient = dJ_sgd(theta, X_b_new[i], y_new[i])
theta = theta - learning_rate(i_iter * m + i) * gradient
return theta
X_b = np.hstack([np.ones((len(X_train), 1)), X_train])
initial_theta = np.random.randn(X_b.shape[1])
self._theta = sgd(X_b, y_train, initial_theta, n_iters, t0, t1)
self.intercept_ = self._theta[0]
self.coef_ = self._theta[1:]
return self
def predict(self, X_predict):
"""给定待预测数据集X_predict,返回表示X_predict的结果向量"""
assert self.intercept_ is not None and self.coef_ is not None, \
"must fit before predict!"
assert X_predict.shape[1] == len(self.coef_), \
"the feature number of X_predict must be equal to X_train"
X_b = np.hstack([np.ones((len(X_predict), 1)), X_predict])
return X_b.dot(self._theta)
def score(self, X_test, y_test):
"""根据测试数据集 X_test 和 y_test 确定当前模型的准确度"""
y_predict = self.predict(X_test)
return r2_score(y_test, y_predict)
def __repr__(self):
return "LinearRegression()"
真实使用我们自己的SGD
model_selection.py
import numpy as np
def train_test_split(X, y, test_ratio=0.2, seed=None):
"""将数据 X 和 y 按照test_ratio分割成X_train, X_test, y_train, y_test"""
assert X.shape[0] == y.shape[0], \
"the size of X must be equal to the size of y"
assert 0.0 <= test_ratio <= 1.0, \
"test_ration must be valid"
if seed:
np.random.seed(seed)
shuffled_indexes = np.random.permutation(len(X))
test_size = int(len(X) * test_ratio)
test_indexes = shuffled_indexes[:test_size]
train_indexes = shuffled_indexes[test_size:]
X_train = X[train_indexes]
y_train = y[train_indexes]
X_test = X[test_indexes]
y_test = y[test_indexes]
return X_train, X_test, y_train, y_test
sklearn中的算法更优
scikit-learn中的SGD
6-8 如何确定梯度计算的准确性 调试梯度下降法
近似计算某点的导数
这样做更容易理解,但其时间复杂度比之前的高
def dJ_debug(theta, X_b, y, epsilon=0.01):
res = np.empty(len(theta))
for i in range(len(theta)):
theta_1 = theta.copy()
theta_1[i] += epsilon
theta_2 = theta.copy()
theta_2[i] -= epsilon
res[i] = (J(theta_1, X_b, y) - J(theta_2, X_b, y)) / (2 * epsilon)
return res
def gradient_descent(dJ, X_b, y, initial_theta, eta, n_iters = 1e4, epsilon=1e-8):
theta = initial_theta
cur_iter = 0
while cur_iter < n_iters:
gradient = dJ(theta, X_b, y)
last_theta = theta
theta = theta - eta * gradient
if(abs(J(theta, X_b, y) - J(last_theta, X_b, y)) < epsilon):
break
cur_iter += 1
return theta
X_b = np.hstack([np.ones((len(X), 1)), X])
initial_theta = np.zeros(X_b.shape[1])
eta = 0.01
%time theta = gradient_descent(dJ_debug, X_b, y, initial_theta, eta)
theta
dJ-debug有作用但速度很慢
dJ-debug的算法与J无法,其它函数都可以用,dJ-math只适用于这一个问题
6-9 有关梯度下降法的更多深入讨论批量,每次看所有的样本
随机每次只看一个
批量慢,在下降最快的方向 随机快但不稳定
将两种方法结合,一次看k个(10或20个)比只看一个的稳定,同时又快
随机的意义
求最小值则加负号,如果求最大值则用正号