MNIST数据集的神经网络推理处理

在上篇博客中，我们学习了mnist数据集下载以及导入的方法，今天尝试用别人训练好的网络权重参数，进行前向网络的搭建，不适用TensorFlow等框架，来体会神经网络各层之间的联系
神经网络权重参数下载链接：（mnist.pkl）https://mp.csdn.net/console/Editorial/14121869
来自《深度学习入门基于python的理论与实现》一书的附赠资源
上一篇博客传送门https://blog.csdn.net/weixin_43872532/article/details/112501990

神经网络的推理处理

神经网络输入层共有784（28*28）个神经元，输出层共有10（0~9共10个取值）个神经元，此外，这个网络共有2个隐藏层，第一个隐藏层有50个神经元，第二个共有100个神经元，其中50和100可以设置成任意值，此处选用书籍提供的权重参数形状50,100
首先定义三个函数get_data(),init_network(),predict()

def get_data():
    (x_train, t_train), (x_test, t_test) = load_mnist(normalize=True, flatten=True, one_hot_label=False)
    return x_test, t_test


def init_network():
    with open("E:\DeepLearning\sample_weight.pkl", 'rb') as f:
        network = pickle.load(f)
    return network


def predict(network, x):
    W1, W2, W3 = network['W1'], network['W2'], network['W3']
    b1, b2, b3 = network['b1'], network['b2'], network['b3']
    
    a1 = np.dot(x, W1) + b1
    z1 = sigmoid(a1)
    a2 = np.dot(z1, W2) + b2
    z2 = sigmoid(a2)
    a3 = np.dot(z2, W3) + b3
    y = softmax(a3)

    return y

其中用到的函数在function.py中有定义（见文末）：
init_network()会读入保存在pickle文件sample_weight.pkl中的权重参数
predict()创建了一个简单的前向神经网络。
接着评价这个网络的识别精度：

x, t = get_data()
network = init_network()
print(network['W3'].shape)
accuracy_cnt = 0
for i in range(len(x)):
    y = predict(network, x[i])
    p= np.argmax(y) # 获取概率最高的元素的索引
    if p == t[i]:
        accuracy_cnt += 1

print("Accuracy:" + str(float(accuracy_cnt) / len(x)))

这就实现了一个简单的神经网络了，整体代码如下：

	# coding: utf-8
import sys, os
import numpy as np
import pickle
from load_mnist import load_mnist
from functions import sigmoid, softmax


def get_data():
    (x_train, t_train), (x_test, t_test) = load_mnist(normalize=True, flatten=True, one_hot_label=False)
    return x_test, t_test


def init_network():
    with open("E:\DeepLearning\sample_weight.pkl", 'rb') as f:
        network = pickle.load(f)
    return network


def predict(network, x):
    W1, W2, W3 = network['W1'], network['W2'], network['W3']
    b1, b2, b3 = network['b1'], network['b2'], network['b3']
    
    a1 = np.dot(x, W1) + b1
    z1 = sigmoid(a1)
    a2 = np.dot(z1, W2) + b2
    z2 = sigmoid(a2)
    a3 = np.dot(z2, W3) + b3
    y = softmax(a3)

    return y


x, t = get_data()
network = init_network()
print(network['W3'].shape)
accuracy_cnt = 0
for i in range(len(x)):
    y = predict(network, x[i])
    p= np.argmax(y) # 获取概率最高的元素的索引
    if p == t[i]:
        accuracy_cnt += 1

print("Accuracy:" + str(float(accuracy_cnt) / len(x))) 

附件

function.py

# coding: utf-8
import numpy as np


def identity_function(x):
    return x


def step_function(x):
    return np.array(x > 0, dtype=np.int)


def sigmoid(x):
    return 1 / (1 + np.exp(-x))    


def sigmoid_grad(x):
    return (1.0 - sigmoid(x)) * sigmoid(x)
    

def relu(x):
    return np.maximum(0, x)


def relu_grad(x):
    grad = np.zeros(x)
    grad[x>=0] = 1
    return grad
    

def softmax(x):
    if x.ndim == 2:
        x = x.T
        x = x - np.max(x, axis=0)
        y = np.exp(x) / np.sum(np.exp(x), axis=0)
        return y.T 

    x = x - np.max(x) # 溢出对策
    return np.exp(x) / np.sum(np.exp(x))


def mean_squared_error(y, t):#均方误差
    return 0.5 * np.sum((y-t)**2)


def cross_entropy_error(y, t):
    if y.ndim == 1:
        t = t.reshape(1, t.size)
        y = y.reshape(1, y.size)
        
    # 监督数据是one-hot-vector的情况下，转换为正确解标签的索引
    if t.size == y.size:
        t = t.argmax(axis=1)
             
    batch_size = y.shape[0]
    return -np.sum(np.log(y[np.arange(batch_size), t] + 1e-7)) / batch_size


def softmax_loss(X, t):
    y = softmax(X)
    return cross_entropy_error(y, t)