机器学习 线性回归算法详解—python代码实现(二)
#%% md
# Logistic Regression
#%% md
## 逻辑回归模型来预测一个学生是否被大学录取
#%% md
我们将建立一个逻辑回归模型来预测一个学生是否被大学录取。假设你是一个大学系的管理员,你想根据两次考试的结果来决定每个申请人的录取机会。你有以前的申请人的历史数据,你可以用它作为逻辑回归的训练集。对于每一个培训例子,你有两个考试的申请人的分数和录取决定。为了做到这一点,我们将建立一个分类模型,根据考试成绩估计入学概率。
#%%
#三大件
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
%matplotlib inline
#%% md
![title](logiReg.jpg)
#%% md
## 范数 np.linalg.norm
![title](范数求解原理.png)
#%%
import os
path = 'LogiReg_data.txt'
pdData = pd.read_csv(path, header=None, names=['Exam 1', 'Exam 2', 'Admitted'])# 由于第一行不是列名,所以header=None, names=[...]
pdData.head()
#%%
pdData.shape # 查看数据的维度
#%%
# 可视化数据的分布状况
positive = pdData[pdData['Admitted'] == 1] # returns the subset of rows such Admitted = 1, i.e. the set of *positive* examples
negative = pdData[pdData['Admitted'] == 0] # returns the subset of rows such Admitted = 0, i.e. the set of *negative* examples
fig, ax = plt.subplots(figsize=(10,5))
ax.scatter(positive['Exam 1'], positive['Exam 2'], s=30, c='b', marker='o', label='Admitted')
ax.scatter(negative['Exam 1'], negative['Exam 2'], s=30, c='r', marker='x', label='Not Admitted')
ax.legend()
ax.set_xlabel('Exam 1 Score')
ax.set_ylabel('Exam 2 Score')
#%% md
## 逻辑回归
#%% md
目标:建立分类器(求解出三个参数 $\theta_0 \theta_1 \theta_2 $)
设定阈值,根据阈值判断录取结果
### 要完成的模块
- `sigmoid` : 映射到概率的函数
- `model` : 返回预测结果值
- `cost` : 根据参数计算损失
- `gradient` : 计算每个参数的梯度方向
- `descent` : 进行参数更新
- `accuracy`: 计算精度
#%% md
### `sigmoid` 函数
$$
g(z) = \frac{1}{1+e^{-z}}
$$
#%%
def sigmoid(z):
return 1 / (1 + np.exp(-z))
#%%
nums = np.arange(-10, 10, step=1) #creates a vector containing 20 equally spaced values from -10 to 10
fig, ax = plt.subplots(figsize=(12,4))
ax.plot(nums, sigmoid(nums), 'r')
#%% md
### Sigmoid
* $g:\mathbb{R} \to [0,1]$
* $g(0)=0.5$
* $g(- \infty)=0$
* $g(+ \infty)=1$
#%%
def model(X, theta):
""" Returns our model result
:param X: examples to classify, n x p
:param theta: parameters, 1 x p
:return: the sigmoid evaluated for each examples in X given parameters theta as a n x 1 vector
"""
return sigmoid(np.dot(X, theta.T))
#%% md
$$
\begin{array}{ccc}
\begin{pmatrix}\theta_{0} & \theta_{1} & \theta_{2}\end{pmatrix} & \times & \begin{pmatrix}1\\
x_{1}\\
x_{2}
\end{pmatrix}\end{array}=\theta_{0}+\theta_{1}x_{1}+\theta_{2}x_{2}
$$
#%%
# 在第0列添加一列数据,值为1,作为偏置x0
pdData.insert(0, 'Ones', 1) # in a try / except structure so as not to return an error if the block si executed several times
# set X (training data) and y (target variable)
orig_data = pdData.values # convert the Pandas representation of the data to an array useful for further computations
orig_data[:10]
#%%
cols = orig_data.shape[1]
X = orig_data[:,0:cols-1] # 即得到样本
y = orig_data[:,cols-1:cols] # 得到标签
# convert to numpy arrays and initalize the parameter array theta
#X = np.matrix(X.values)
#y = np.matrix(data.iloc[:,3:4].values) #np.array(y.values)
theta = np.zeros([1, 3])
#%%
X[:5]
#%%
y[:5]
#%%
theta
#%%
X.shape, y.shape, theta.shape
#%% md
### 损失函数
将对数似然函数去负号
$$
D(h_\theta(x), y) = -y\log(h_\theta(x)) - (1-y)\log(1-h_\theta(x))
$$
求平均损失
$$
J(\theta)=\frac{1}{n}\sum_{i=1}^{n} D(h_\theta(x_i), y_i)
$$
#%%
def cost(X, y, theta):
left = np.multiply(-y, np.log(model(X, theta)))
right = np.multiply(1 - y, np.log(1 - model(X, theta)))
return np.sum(left - right) / (len(X))
#%%
cost(X, y, theta)
#%% md
### 计算梯度
$$
\frac{\partial J}{\partial \theta_j}=-\frac{1}{m}\sum_{i=1}^n (y_i - h_\theta (x_i))x_{ij}
$$
#%%
def gradient(X, y, theta):
grad = np.zeros(theta.shape) # 梯度与theta的维度保持一致
error = (model(X, theta)- y).ravel()
for j in range(len(theta.ravel())): #for each parmeter
term = np.multiply(error, X[:,j])
grad[0, j] = np.sum(term) / len(X)
return grad # 返回3个方向的梯度
#%% md
### Gradient descent
#%% md
比较3中不同梯度下降方法
#%%
STOP_ITER = 0 # 根据迭代轮数停止迭代
STOP_COST = 1 # 根据损失停止迭代
STOP_GRAD = 2 # 根据梯度停止迭代
def stopCriterion(type, value, threshold):
#设定三种不同的停止策略
if type == STOP_ITER: return value > threshold
elif type == STOP_COST: return abs(value[-1]-value[-2]) < threshold
elif type == STOP_GRAD: return np.linalg.norm(value) < threshold # 求得的是范数(此处默认是对整体矩阵求范数)
#%%
import numpy.random
#洗牌
def shuffleData(data):
np.random.shuffle(data)
cols = data.shape[1]
X = data[:, 0:cols-1]
y = data[:, cols-1:]
return X, y
#%%
import time
# 梯度下降求解
def descent(data, theta, batchSize, stopType, thresh, alpha):
# batchSize:如果为1,则为随机梯度下降;如果指定为样本数,则为梯度下降;如果小于样本数,则为小批次梯度下降
# stopType:停止策略
# thresh:策略对应的阈值
# alpha:学习率
init_time = time.time()
i = 0 # 迭代次数
k = 0 # batch
X, y = shuffleData(data) # 数据打乱
grad = np.zeros(theta.shape) # 计算的梯度
costs = [cost(X, y, theta)] # 计算损失值
while True:
grad = gradient(X[k:k+batchSize], y[k:k+batchSize], theta) # 计算梯度
k += batchSize #取batch数量个数据
if k >= n:
k = 0
X, y = shuffleData(data) #重新洗牌
theta = theta - alpha*grad # 进行参数更新(即进行梯度下降操作)
costs.append(cost(X, y, theta)) # 计算新的损失
i += 1
if stopType == STOP_ITER: value = i
elif stopType == STOP_COST: value = costs
elif stopType == STOP_GRAD: value = grad
if stopCriterion(stopType, value, thresh): break
return theta, i-1, costs, grad, time.time() - init_time
#%%
def runExpe(data, theta, batchSize, stopType, thresh, alpha):
# batchSize:如果为1,则为随机梯度下降;如果指定为样本数,则为梯度下降;如果小于样本数,则为小批次梯度下降
# stopType:停止策略
# thresh:阈值
# alpha:学习率
#import pdb; pdb.set_trace();
# 执行梯度下降算法
theta, iter, costs, grad, dur = descent(data, theta, batchSize, stopType, thresh, alpha)
name = "Original" if (data[:,1]>2).sum() > 1 else "Scaled"
name += " data - learning rate: {} - ".format(alpha)
# 梯度下降方式
if batchSize==n: strDescType = "Gradient"
elif batchSize==1: strDescType = "Stochastic"
else: strDescType = "Mini-batch ({})".format(batchSize)
name += strDescType + " descent - Stop: "
# 停止方式
if stopType == STOP_ITER: strStop = "{} iterations".format(thresh)
elif stopType == STOP_COST: strStop = "costs change < {}".format(thresh)
else: strStop = "gradient norm < {}".format(thresh)
name += strStop
print ("***{}\nTheta: {} - Iter: {} - Last cost: {:03.2f} - Duration: {:03.2f}s".format(
name, theta, iter, costs[-1], dur))
fig, ax = plt.subplots(figsize=(12,4))
ax.plot(np.arange(len(costs)), costs, 'r')
ax.set_xlabel('Iterations')
ax.set_ylabel('Cost')
ax.set_title(name.upper() + ' - Error vs. Iteration')
return theta
#%% md
### 不同的停止策略
#%% md
#### 设定迭代次数
#%%
#选择的梯度下降方法是基于所有样本的
n=100 # 批次
runExpe(orig_data, theta, n, STOP_ITER, thresh=5000, alpha=0.000001)
#%% md
#### 根据损失值停止
#%% md
设定阈值 1E-6, 差不多需要110 000次迭代
#%%
runExpe(orig_data, theta, n, STOP_COST, thresh=0.000001, alpha=0.001)
#%% md
#### 根据梯度变化停止
#%% md
设定阈值 0.05,差不多需要40 000次迭代
#%%
runExpe(orig_data, theta, n, STOP_GRAD, thresh=0.05, alpha=0.001)
#%% md
### 对比不同的梯度下降方法
#%% md
#### Stochastic descent
#%%
runExpe(orig_data, theta, 1, STOP_ITER, thresh=5000, alpha=0.001) # 即使用随机梯度下降
#%% md
有点爆炸。。。很不稳定,再来试试把学习率调小一些
#%%
runExpe(orig_data, theta, 1, STOP_ITER, thresh=15000, alpha=0.000002) # 即使用随机梯度下降
#%% md
速度快,但稳定性差,需要很小的学习率
#%% md
#### Mini-batch descent
#%%
runExpe(orig_data, theta, 16, STOP_ITER, thresh=15000, alpha=0.001) # 小批次梯度下降
#%% md
浮动仍然比较大,我们来尝试下对数据进行标准化
将数据按其属性(按列进行)减去其均值,然后除以其方差。最后得到的结果是,对每个属性/每列来说所有数据都聚集在0附近,方差值为1
#%%
# 使用sklearn将数据标准化处理,即对数据减去均值,在除以其标准差
from sklearn import preprocessing as pp
scaled_data = orig_data.copy()
scaled_data[:, 1:3] = pp.scale(orig_data[:, 1:3])
runExpe(scaled_data, theta, n, STOP_ITER, thresh=5000, alpha=0.001)
#%% md
它好多了!原始数据,只能达到达到0.61,而我们得到了0.38个在这里!
所以对数据做预处理是非常重要的
#%%
runExpe(scaled_data, theta, n, STOP_GRAD, thresh=0.02, alpha=0.001)
#%% md
更多的迭代次数会使得损失下降的更多!
#%%
theta = runExpe(scaled_data, theta, 1, STOP_GRAD, thresh=0.002/5, alpha=0.001)
#%% md
随机梯度下降更快,但是我们需要迭代的次数也需要更多,所以还是用batch的比较合适!!!
#%%
runExpe(scaled_data, theta, 16, STOP_GRAD, thresh=0.002*2, alpha=0.001)
#%% md
## 精度
#%%
#设定阈值(即将概率转换成一个类别)
def predict(X, theta):
return [1 if x >= 0.5 else 0 for x in model(X, theta)]
#%%
scaled_X = scaled_data[:, :3]
y = scaled_data[:, 3]
predictions = predict(scaled_X, theta)
correct = [1 if ((a == 1 and b == 1) or (a == 0 and b == 0)) else 0 for (a, b) in zip(predictions, y)]
accuracy = (sum(map(int, correct)) % len(correct))
print ('accuracy = {0}%'.format(accuracy))
#%%