main.py
import random
import numpy as np
import matplotlib.pyplot as plt
def load_dataset():
"""
前两行两个值分别是X1和X2,第三个是标签类别
"""
data_mat = []
label_mat = []
# 读取格式:-0.017612 14.053064 0
with open('testSet.txt', 'r') as f:
for line in f.readlines():
# 拆分放到arr里面
line_arr = line.strip().split()
# 第一位是1?
data_mat.append([1., float(line_arr[0]), float(line_arr[1])])
# 标签变成2
label_mat.append(int(line_arr[2]))
return data_mat, label_mat
def sigmoid(inx):
# 那个s形的函数
return 1.0 / (1 + np.exp(-inx))
def grad_ascent(data_mat, class_labels):
"""
梯度上升
:param data_mat:
:param class_labels:
:return:
"""
# matrix格式
data_matrix = np.mat(data_mat) # 2D np array,列是特征,行是样本
# 也是matrix格式,并站了起来
label_mat = np.mat(class_labels).transpose() # 转置,行变列
m, n = np.shape(data_matrix)
alpha = 0.001 # lr
max_cycles = 500 # iter
weights = np.ones((n, 1))
# 保存weight
weights_history = np.zeros((max_cycles, n))
for k in range(max_cycles):
# alpha = 0.004 / (1.0 + k) + 0.001
h = sigmoid(data_matrix * weights) # 是一个向量
error = (label_mat - h) # 真是简陋
# 每次都撸一个进去
weights = weights + alpha * data_matrix.transpose() * error
weights_history[k, :] = weights.transpose()
# 根本就不收敛
return weights, weights_history
def stoc_grad_ascent(data_mat, class_labels, iter_counts=20):
m, n = np.shape(data_mat)
# alpha = 0.01 # 改进前
weights = np.ones(n)
# 迭代次数
# iter_counts = 20
# 存放历史
weights_history = np.zeros((m * iter_counts, n))
# 只跑那么多样本
for j in range(iter_counts):
data_index = list(range(m))
for i in range(m):
# 模拟退火常见?
alpha = 4 / (1.0 + j + i) + 0.01 # 改进后,减少波动
# 下面这种改动似乎不靠谱
# rand_index = int(random.uniform(0,len(data_index)))
h = sigmoid(np.sum(data_mat[i] * weights))
error = class_labels[i] - h
weights = weights + alpha * error * data_mat[i]
weights_history[i + j * m, :] = weights
# data_index.pop(rand_index)
return weights, weights_history
def plot_data(weights):
"""
画决策边界和散点图
:param weights:
:return:
"""
import matplotlib.pylab as plt
# 原本的weights是matrix,要转换为ndarray
weights = np.squeeze(np.asarray(weights))
data_mat, label_mat = load_dataset()
data_arr = np.array(data_mat)
n = np.shape(data_arr)[0]
xcord1, ycord1, xcord2, ycord2 = [], [], [], []
for i in range(n):
# 分两类
if int(label_mat[i]) == 1:
xcord1.append(data_arr[i, 1])
ycord1.append(data_arr[i, 2])
else:
xcord2.append(data_arr[i, 1])
ycord2.append(data_arr[i, 2])
# 创建画板
fig = plt.figure()
# 设置布局
ax = fig.add_subplot(111)
# 散点图
ax.scatter(xcord1, ycord1, s=30, c='red', marker='s')
ax.scatter(xcord2, ycord2, s=30, c='green')
x = np.arange(-3.0, 3.0, 0.1)
y = (-weights[0] - weights[1] * x) / weights[2]
# 连续的线段
ax.plot(x, y)
plt.xlabel('X1')
plt.ylabel('X2')
plt.show()
def plot_weights_update(weights_history):
"""用来画权重的更新图"""
fig = plt.figure()
# 三行一列的第一行
ax = fig.add_subplot(311)
type1 = ax.plot(weights_history[:, 0])
plt.ylabel('X0')
ax = fig.add_subplot(312)
type2 = ax.plot(weights_history[:, 1])
plt.ylabel('X1')
ax = fig.add_subplot(313)
type3 = ax.plot(weights_history[:, 2])
plt.xlabel('iteration')
plt.ylabel('X2')
plt.show()
if __name__ == '__main__':
data_arr, label_mat = load_dataset() # 加载数据
weights1, weights_history1 = grad_ascent(data_arr, label_mat) # 梯度上升
weights, weights_history = stoc_grad_ascent(np.array(data_arr), label_mat) # 随机梯度
plot_weights_update(weights_history) # 随机
plot_weights_update(weights_history1) # 梯度
plot_data(weights) # 随机
plot_data(weights1) # 梯度
pass
梯度上升
500次梯度上升:
梯度上升的收敛结果:
随机梯度上升
一次随机梯度上升:
与之对应的一次weights学习图:
对随机梯度迭代20次,并加入学习率衰减,与梯度上升结果基本一样:
再看收敛图:
完美
附录:
testSet.txt
-0.017612 14.053064 0
-1.395634 4.662541 1
-0.752157 6.538620 0
-1.322371 7.152853 0
0.423363 11.054677 0
0.406704 7.067335 1
0.667394 12.741452 0
-2.460150 6.866805 1
0.569411 9.548755 0
-0.026632 10.427743 0
0.850433 6.920334 1
1.347183 13.175500 0
1.176813 3.167020 1
-1.781871 9.097953 0
-0.566606 5.749003 1
0.931635 1.589505 1
-0.024205 6.151823 1
-0.036453 2.690988 1
-0.196949 0.444165 1
1.014459 5.754399 1
1.985298 3.230619 1
-1.693453 -0.557540 1
-0.576525 11.778922 0
-0.346811 -1.678730 1
-2.124484 2.672471 1
1.217916 9.597015 0
-0.733928 9.098687 0
-3.642001 -1.618087 1
0.315985 3.523953 1
1.416614 9.619232 0
-0.386323 3.989286 1
0.556921 8.294984 1
1.224863 11.587360 0
-1.347803 -2.406051 1
1.196604 4.951851 1
0.275221 9.543647 0
0.470575 9.332488 0
-1.889567 9.542662 0
-1.527893 12.150579 0
-1.185247 11.309318 0
-0.445678 3.297303 1
1.042222 6.105155 1
-0.618787 10.320986 0
1.152083 0.548467 1
0.828534 2.676045 1
-1.237728 10.549033 0
-0.683565 -2.166125 1
0.229456 5.921938 1
-0.959885 11.555336 0
0.492911 10.993324 0
0.184992 8.721488 0
-0.355715 10.325976 0
-0.397822 8.058397 0
0.824839 13.730343 0
1.507278 5.027866 1
0.099671 6.835839 1
-0.344008 10.717485 0
1.785928 7.718645 1
-0.918801 11.560217 0
-0.364009 4.747300 1
-0.841722 4.119083 1
0.490426 1.960539 1
-0.007194 9.075792 0
0.356107 12.447863 0
0.342578 12.281162 0
-0.810823 -1.466018 1
2.530777 6.476801 1
1.296683 11.607559 0
0.475487 12.040035 0
-0.783277 11.009725 0
0.074798 11.023650 0
-1.337472 0.468339 1
-0.102781 13.763651 0
-0.147324 2.874846 1
0.518389 9.887035 0
1.015399 7.571882 0
-1.658086 -0.027255 1
1.319944 2.171228 1
2.056216 5.019981 1
-0.851633 4.375691 1
-1.510047 6.061992 0
-1.076637 -3.181888 1
1.821096 10.283990 0
3.010150 8.401766 1
-1.099458 1.688274 1
-0.834872 -1.733869 1
-0.846637 3.849075 1
1.400102 12.628781 0
1.752842 5.468166 1
0.078557 0.059736 1
0.089392 -0.715300 1
1.825662 12.693808 0
0.197445 9.744638 0
0.126117 0.922311 1
-0.679797 1.220530 1
0.677983 2.556666 1
0.761349 10.693862 0
-2.168791 0.143632 1
1.388610 9.341997 0
0.317029 14.739025 0