Logistic回归(分类问题)
这部分教程将介绍一部分:
- Logistic分类模型
我们在上次的教程中给出了一个很简单的模型,只有一个输入和一个输出。在这篇教程中,我们将构建一个二分类模型,输入参数是两个变量。这个模型在统计上被称为Logistic回归模型,网络结构可以被描述如下:
![](https://i-blog.csdnimg.cn/blog_migrate/299227b19e2faccbe0f837aa8a20e438.webp?x-image-process=image/format,png)
我们先导入教程需要使用的软件包。
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.colors import colorConverter, ListedColormap
from matplotlib import cm
定义类分布
在教程中,目标分类t
将从两个独立分布中产生,当t=1
时,用蓝色表示。当t=0
时,用红色表示。输入参数X
是一个N*2
的矩阵,目标分类t
是一个N * 1
的向量。更直观的表现,见下图。
# Define and generate the samples
nb_of_samples_per_class = 20 # The number of sample in each class
red_mean = [-1,0] # The mean of the red class
blue_mean = [1,0] # The mean of the blue class
std_dev = 1.2 # standard deviation of both classes
# Generate samples from both classes
x_red = np.random.randn(nb_of_samples_per_class, 2) * std_dev + red_mean
x_blue = np.random.randn(nb_of_samples_per_class, 2) * std_dev + blue_mean
# Merge samples in set of input variables x, and corresponding set of output variables t
X = np.vstack((x_red, x_blue))
t = np.vstack((np.zeros((nb_of_samples_per_class,1)), np.ones((nb_of_samples_per_class,1))))
# Plot both classes on the x1, x2 plane
plt.plot(x_red[:,0], x_red[:,1], 'ro', label='class red')
plt.plot(x_blue[:,0], x_blue[:,1], 'bo', label='class blue')
plt.grid()
plt.legend(loc=2)
plt.xlabel('$x_1$', fontsize=15)
plt.ylabel('$x_2$', fontsize=15)
plt.axis([-4, 4, -4, 4])
plt.title('red vs. blue classes in the input space')
plt.show()
![](https://i-blog.csdnimg.cn/blog_migrate/17316439baa09fd20b7998d070e80d6e.webp?x-image-process=image/format,png)
Logistic函数和交叉熵损失函数
Logistic函数
我们设计的网络的目的是从输入的x
去预测目标t
。假设,输入x = [x1, x2]
,权重w = [w1, w2]
,预测目标t = 1
。那么,概率P(t = 1|x, w)
将是神经网络输出的y
,即y = σ(x∗wT)
。其中,σ
表示Logistic函数
,定义如下:
![](https://i-blog.csdnimg.cn/blog_migrate/c02ec42c4e7f9ddd82c8be5975aa3a6c.webp?x-image-process=image/format,png)
如果,对于Logistic函数和它的导数还不是很清楚的,可以查看这个教程,里面进行了详细描述。
交叉熵损失函数
对于这个分类问题的损失函数优化,我们使用交叉熵误差函数来解决,对于每个训练样本i
,交叉熵误差函数定义如下:
![](https://i-blog.csdnimg.cn/blog_migrate/582df437c6db00f9613ca81816d5be87.webp?x-image-process=image/format,png)
i
的交叉熵误差函数
如果我们要计算整个训练样本的交叉熵误差,那么只需要把每一个样本的值进行累加就可以了,即:
![](https://i-blog.csdnimg.cn/blog_migrate/7725572e001856215036ab7a9347e379.webp?x-image-process=image/format,png)
关于交叉熵误差函数更加详细的介绍可以看这个教程。
logistic(z)
函数实现了Logistic
函数,cost(y, t)
函数实现了损失函数,nn(x, w)
实现了神经网络的输出结果,nn_predict(x, w)
实现了神经网络的预测结果。
# Define the logistic function
def logistic(z):
return 1 / (1 + np.exp(-z))
# Define the neural network function y = 1 / (1 + numpy.exp(-x*w))
def nn(x, w):
return logistic(x.dot(w.T))
# Define the neural network prediction function that only returns
# 1 or 0 depending on the predicted class
def nn_predict(x,w):
return np.around(nn(x,w))
# Define the cost function
def cost(y, t):
return - np.sum(np.multiply(t, np.log(y)) + np.multiply((1-t), np.log(1-y)))
# Plot the cost in function of the weights
# Define a vector of weights for which we want to plot the cost
nb_of_ws = 100 # compute the cost nb_of_ws times in each dimension
ws1 = np.linspace(-5, 5, num=nb_of_ws) # weight 1
ws2 = np.linspace(-5, 5, num=nb_of_ws) # weight 2
ws_x, ws_y = np.meshgrid(ws1, ws2) # generate grid
cost_ws = np.zeros((nb_of_ws, nb_of_ws)) # initialize cost matrix
# Fill the cost matrix for each combination of weights
for i in range(nb_of_ws):
for j in range(nb_of_ws):
cost_ws[i,j] = cost(nn(X, np.asmatrix([ws_x[i,j], ws_y[i,j]])) , t)
# Plot the cost function surface
plt.contourf(ws_x, ws_y, cost_ws, 20, cmap=cm.pink)
cbar = plt.colorbar()
cbar.ax.set_ylabel('$\\xi$', fontsize=15)
plt.xlabel('$w_1$', fontsize=15)
plt.ylabel('$w_2$', fontsize=15)
plt.title('Cost function surface')
plt.grid()
plt.show()
![](https://i-blog.csdnimg.cn/blog_migrate/31ec8f107221f736237e925c7ebdda61.webp?x-image-process=image/format,png)
梯度下降优化损失函数
梯度下降算法的工作原理是损失函数ξ
对于每一个参数的求导,然后沿着负梯度方向进行参数更新。
参数w
按照一定的学习率沿着负梯度方向更新,即w(k+1)=w(k)−Δw(k+1)
,其中Δw
可以表示为:
![](https://i-blog.csdnimg.cn/blog_migrate/b45f064df9c56ae34b36d1543e122efb.webp?x-image-process=image/format,png)
对于每个训练样本i
,∂ξi/∂w
计算如下:
![](https://i-blog.csdnimg.cn/blog_migrate/34fd456f2843f26324b830eac3eba960.webp?x-image-process=image/format,png)
其中,yi=σ(zi)
是神经元的Logistic
输出,zi=xi∗wT
是神经元的输入。
在详细推导损失函数对于权重的导数之前,我们先这个教程中摘取几个推导。
![](https://i-blog.csdnimg.cn/blog_migrate/d93732d5d598ce09c7bf7d9288740b1f.webp?x-image-process=image/format,png)
参考上面的分步推导,我们可以得到下面的详细推导:
![](https://i-blog.csdnimg.cn/blog_migrate/207d78de967483d07985d934c91811ac.webp?x-image-process=image/format,png)
因此,对于每个权重的更新Δwj
可以表示为:
![](https://i-blog.csdnimg.cn/blog_migrate/d22be998398c0373211494f10d021ef4.webp?x-image-process=image/format,png)
在批处理中,我们需要将N
个样本的梯度都进行累加,即:
![](https://i-blog.csdnimg.cn/blog_migrate/380a72578756490231c11c222e79e89b.webp?x-image-process=image/format,png)
在开始梯度下降算法之前,你需要对参数都进行一个随机数赋值过程,然后采用梯度下降算法更新参数,直至收敛。
gradient(w, x, t)
函数实现了梯度∂ξ/∂w
,delta_w(w_k, x, t, learning_rate)
函数实现了Δw
。
# define the gradient function.
def gradient(w, x, t):
return (nn(x, w) - t).T * x
# define the update function delta w which returns the
# delta w for each weight in a vector
def delta_w(w_k, x, t, learning_rate):
return learning_rate * gradient(w_k, x, t)
梯度下降更新
我们在训练集X
上面运行10
次去做预测,下图中画出了前三次的结果,图中蓝色的点表示在第k
次,w(k)
的值。
# Set the initial weight parameter
w = np.asmatrix([-4, -2])
# Set the learning rate
learning_rate = 0.05
# Start the gradient descent updates and plot the iterations
nb_of_iterations = 10 # Number of gradient descent updates
w_iter = [w] # List to store the weight values over the iterations
for i in range(nb_of_iterations):
dw = delta_w(w, X, t, learning_rate) # Get the delta w update
w = w-dw # Update the weights
w_iter.append(w) # Store the weights for plotting
# Plot the first weight updates on the error surface
# Plot the error surface
plt.contourf(ws_x, ws_y, cost_ws, 20, alpha=0.9, cmap=cm.pink)
cbar = plt.colorbar()
cbar.ax.set_ylabel('cost')
# Plot the updates
for i in range(1, 4):
w1 = w_iter[i-1]
w2 = w_iter[i]
# Plot the weight-cost value and the line that represents the update
plt.plot(w1[0,0], w1[0,1], 'bo') # Plot the weight cost value
plt.plot([w1[0,0], w2[0,0]], [w1[0,1], w2[0,1]], 'b-')
plt.text(w1[0,0]-0.2, w1[0,1]+0.4, '$w({})$'.format(i), color='b')
w1 = w_iter[3]
# Plot the last weight
plt.plot(w1[0,0], w1[0,1], 'bo')
plt.text(w1[0,0]-0.2, w1[0,1]+0.4, '$w({})$'.format(4), color='b')
# Show figure
plt.xlabel('$w_1$', fontsize=15)
plt.ylabel('$w_2$', fontsize=15)
plt.title('Gradient descent updates on cost surface')
plt.grid()
plt.show()
![](https://i-blog.csdnimg.cn/blog_migrate/ada2bdc2215f48b7480ef8bf47d56a0e.webp?x-image-process=image/format,png)
训练结果可视化
下列代码,我们将训练的结果进行可视化。
# Plot the resulting decision boundary
# Generate a grid over the input space to plot the color of the
# classification at that grid point
nb_of_xs = 200
xs1 = np.linspace(-4, 4, num=nb_of_xs)
xs2 = np.linspace(-4, 4, num=nb_of_xs)
xx, yy = np.meshgrid(xs1, xs2) # create the grid
# Initialize and fill the classification plane
classification_plane = np.zeros((nb_of_xs, nb_of_xs))
for i in range(nb_of_xs):
for j in range(nb_of_xs):
classification_plane[i,j] = nn_predict(np.asmatrix([xx[i,j], yy[i,j]]) , w)
# Create a color map to show the classification colors of each grid point
cmap = ListedColormap([
colorConverter.to_rgba('r', alpha=0.30),
colorConverter.to_rgba('b', alpha=0.30)])
# Plot the classification plane with decision boundary and input samples
plt.contourf(xx, yy, classification_plane, cmap=cmap)
plt.plot(x_red[:,0], x_red[:,1], 'ro', label='target red')
plt.plot(x_blue[:,0], x_blue[:,1], 'bo', label='target blue')
plt.grid()
plt.legend(loc=2)
plt.xlabel('$x_1$', fontsize=15)
plt.ylabel('$x_2$', fontsize=15)
plt.title('red vs. blue classification boundary')
plt.show()
![](https://i-blog.csdnimg.cn/blog_migrate/fccbc1267df029a54e11ce947dcd7288.webp?x-image-process=image/format,png)