手写数字识别单一图像和批量图像—python实现

手写数字识别 python实现

# coding: utf-8
import sys, os
sys.path.append(os.pardir) 
import numpy as np
import pickle
from dataset.mnist import load_mnist
import time

def softmax(x):
    x = x - np.max(x) # 溢出对策
    return np.exp(x) / np.sum(np.exp(x))
def sigmoid(x):
    return 1 / (1 + np.exp(-x))  

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("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()
accuracy_cnt = 0
start = time.time()
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
end = time.time()
print("time:",end-start)
print("Accuracy:" + str(float(accuracy_cnt) / len(x)))

time: 1.3383796215057373
Accuracy:0.9352

网络形状解析

    x,_ = get_data()
    print(x.shape)
    W1, W2, W3 = network['W1'], network['W2'], network['W3']
    print(W1.shape,W2.shape,W3.shape)

输出结果如下：

(10000, 784)#测试数据为10000 784
(784, 50) (50, 100) (100, 10)

处理过程：输入一个由784个元素构成的一维数组后，输出一个有10个元素的一维数组。
如果输入多张呢？

输入数据的形状为 100 × 784，输出数据的形状为
100 × 10。这表示输入的100张图像的结果被一次性输出了
批处理对计算机的运算大有好处，可以缩短每张图像的处理时间。因为大多数处理数值计算的库都进行了能够高效处理大型数组运算的最优化。并且在神经网络中，当数据传送称为瓶颈时，可以将更多的时间用在计算上。

# coding: utf-8
import sys, os
sys.path.append(os.pardir)  
import numpy as np
import pickle
from dataset.mnist import load_mnist

def softmax(x):
    x = x - np.max(x) # 溢出对策
    return np.exp(x) / np.sum(np.exp(x))
def sigmoid(x):
    return 1 / (1 + np.exp(-x))  

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("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()

batch_size = 100 # 批数量
accuracy_cnt = 0
start = time.time()
for i in range(0, len(x), batch_size):
    x_batch = x[i:i+batch_size]
    y_batch = predict(network, x_batch)
    p = np.argmax(y_batch, axis=1)
    accuracy_cnt += np.sum(p == t[i:i+batch_size])
end = time.time()
print("Accuracy:" + str(float(accuracy_cnt) / len(x)))

time: 0.06778287887573242
Accuracy:0.9352

批量处理时间比不批量快大约20倍