深度学习入门初探——损失函数均方误差

        前文已经介绍了损失函数的目的是做什么了的：

        深度学习入门初探——引出损失函数https://www.toutiao.com/i7033404064663126560/?group_id=7033404064663126560

        这里呢，先来学习下均方误差这个损失函数是怎么计算的：

        y表示测试数据经过预测得到的结果，t表示测试数据的正确结果，k表示数据的维度。下面是python实现的代码：

def maen_squared_error(r1, r2):
    return 0.5*np.sum((r1-r2)**2)

---

        为了更好地学习这个损失函数的指标性质，在这里拿深度学习入门初探——hello word式的手写数字识别中的数据做个演示：很明显可以看出那个数据吻合的更好。

# coding: utf-8
import sys, os
sys.path.append(os.pardir)
import numpy as np
import pickle
from common.functions import sigmoid, softmax

def _change_one_hot_label(X):
    T = np.zeros((X.size, 10))
    for idx, row in enumerate(T):
        row[X[idx]] = 1
    return T
    
def load_mnist(normalize=True, flatten=True, one_hot_label=False):
    """读入MNIST数据集
    
    Parameters
    ----------
    normalize : 将图像的像素值正规化为0.0~1.0
    one_hot_label : 
        one_hot_label为True的情况下，标签作为one-hot数组返回
        one-hot数组是指[0,0,1,0,0,0,0,0,0,0]这样的数组
    flatten : 是否将图像展开为一维数组
    
    Returns
    -------
    (训练图像, 训练标签), (测试图像, 测试标签)
    """        
    with open("mnist.pkl", 'rb') as f:
        dataset = pickle.load(f)
    
    if normalize:
        for key in ('train_img', 'test_img'):
            dataset[key] = dataset[key].astype(np.float32)
            dataset[key] /= 255.0
            
    if one_hot_label:
        dataset['train_label'] = _change_one_hot_label(dataset['train_label'])
        dataset['test_label'] = _change_one_hot_label(dataset['test_label'])
    
    if not flatten:
         for key in ('train_img', 'test_img'):
            dataset[key] = dataset[key].reshape(-1, 1, 28, 28)

    return (dataset['train_img'], dataset['train_label']), (dataset['test_img'], dataset['test_label']) 

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

def maen_squared_error(r1, r2):
    return 0.5*np.sum((r1-r2)**2)


np.set_printoptions(formatter={'float': '{: 0.9f}'.format})
network = init_network()
x, t = get_data()

y = predict(network, x[10])
print("\n数据10预测的输出:\n",y, "\n")
print("数据10实际的结果:\n",t[10], "\n")
msr0 = maen_squared_error(y, t[10])
print("数据10预测输出与实际结果的均方误差:",msr0, "\n")

y = predict(network, x[1])
print("数据1预测的输出:\n",y, "\n")
print("数据1实际的结果:\n",t[1], "\n")
msr1 = maen_squared_error(y, t[1])
print("数据0预测输出与实际结果的均方误差:",msr1, "\n")