NumPy

导包

import numpy as np
import matplotlib.pyplot as plt
import h5py
import scipy
from PIL import Image
from scipy import ndimage
from lr_utils import load_dataset

%matplotlib inline

加载数据

train_set_x_orig, train_set_y, test_set_x_orig, test_set_y, classes = load_dataset()

m_train = train_set_x_orig.shape[0]
m_test = test_set_x_orig.shape[0]
num_px = test_set_x_orig.shape[1]

转换数据的形状，转换为：每一列代表一个图片的RGB，行代表所有的图片

train_set_x_flatten = train_set_x_orig.reshape(train_set_x_orig.shape[0], -1).T
test_set_x_flatten = test_set_x_orig.reshape(test_set_x_orig.shape[0], -1).T
print(train_set_x_flatten.shape)
print(test_set_x_flatten.shape)

(12288, 209)
(12288, 50)

将数据集标准化（颜色值都是0-255，要把颜色值转化成0-1）

train_set_x = train_set_x_flatten/255
test_set_x = test_set_x_flatten/255

sigmoid 函数

def sigmoid(z):
    s = 1/(1+np.exp(-z))
    return s

初始化w、b(这里是logistics regression)

def initialize_with_zeros(dim):
    w = np.zeros(shape=(dim,1))
    b = 0
    assert(w.shape == (dim, 1)) #断言确保维度正确
    assert(isinstance(b, float) or isinstance(b, int))
    
    return w,b

向前向后传播

def propagate(w, b, X, Y):
    #向前传播
    m = X.shape[1] #样本个数
    A = sigmoid(np.dot(w.T, X)+b)
    cost = (-1/m)*np.sum(Y*np.log(A)+(1-Y)*np.log(1-A))
    
    #向后传播
    dw = 1/m * np.dot(X , (A - Y).T)
    db = 1/m * np.sum(A - Y)
    
    assert(dw.shape == w.shape)
    assert(db.dtype == float)
    cost = np.squeeze(cost)
    assert(cost.shape == ())
    
    grads = {"dw": dw,  #将dw、db存入字典
            "db": db}
    
    return grads,cost

优化w、b

def optimize(w, b, X, Y, num_iterations, learning_rate, print_cost=False):
    """
    num_iterations  - 优化循环的迭代次数
        learning_rate  - 梯度下降更新规则的学习率
        print_cost  - 每100步打印一次损失值
    """
    costs = []
    for i in range(num_iterations):
        grads, cost = propagate(w,b,X,Y)
        
        dw = grads["dw"]
        db = grads["db"]
        
        w = w-learning_rate*dw
        b = b-learning_rate*db
        
        if i%100 == 0:
            costs.append(cost)
        if print_cost and i % 100 == 0:
            print("Cost after iteration %i: %f" %(i, cost))
        
    params = {"w": w,
             "b": b}
    
    grads = {"dw": dw,
            "db": db}
    
    return params,grads,costs

预测

def predict(w, b, X):
    m = X.shape[1]
    Y_prediction = np.zeros((1,m)) #每一列对应一张图片
    w = w.reshape(X.shape[0], 1) #X.shape[0]代表一章图片的每一个颜色值，要和w权值相乘
    
    A = sigmoid(np.dot(w.T, X)+b) #激活函数将值对应在0-1（logistics regression）
    
    for i in range(A.shape[1]):
        if(A[0,i]>0.5):
            Y_prediction[0,i]=1
        else:
            Y_prediction[0,i]=0
    assert(Y_prediction.shape == (1, m))
    return Y_prediction,A

制作模型

def model(X_train, Y_train, X_test, Y_test, num_iterations = 2000, learning_rate = 0.5, print_cost = False):
    """
    num_iterations  - 优化循环的迭代次数
        learning_rate  - 梯度下降更新规则的学习率
        print_cost  - 每100步打印一次损失值
    """
    # 初始化
    w, b = initialize_with_zeros(X_train.shape[0])
    
    parameters, grads, costs = optimize(w, b, X_train, Y_train, num_iterations, learning_rate, print_cost)
    
    w = parameters["w"]
    b = parameters["b"]
    
    Y_prediction_test, A_test = predict(w, b, X_test)
    Y_prediction_train, A_train = predict(w, b, X_train)
    
    #打印错误预测
    # Print train/test Errors
    print("train accuracy: {} %".format(100 - np.mean(np.abs(Y_prediction_train - Y_train)) * 100))
    print("test accuracy: {} %".format(100 - np.mean(np.abs(Y_prediction_test - Y_test)) * 100))
    
    d = {"costs": costs,
         "Y_prediction_test": Y_prediction_test, 
         "Y_prediction_train" : Y_prediction_train, 
         "w" : w, 
         "b" : b,
         "learning_rate" : learning_rate,
         "num_iterations": num_iterations}
    
    return d

d = model(train_set_x, train_set_y, test_set_x, test_set_y, num_iterations = 2000, learning_rate = 0.004, print_cost = True)

Cost after iteration 0: 0.693147
Cost after iteration 100: 0.506765
Cost after iteration 200: 0.442269
Cost after iteration 300: 0.397201
Cost after iteration 400: 0.362439
Cost after iteration 500: 0.334271
Cost after iteration 600: 0.310725
Cost after iteration 700: 0.290608
Cost after iteration 800: 0.273138
Cost after iteration 900: 0.257771
Cost after iteration 1000: 0.244114
Cost after iteration 1100: 0.231873
Cost after iteration 1200: 0.220823
Cost after iteration 1300: 0.210787
Cost after iteration 1400: 0.201623
Cost after iteration 1500: 0.193217
Cost after iteration 1600: 0.185474
Cost after iteration 1700: 0.178317
Cost after iteration 1800: 0.171678
Cost after iteration 1900: 0.165504
train accuracy: 98.08612440191388 %
test accuracy: 70.0 %

# 学习速率对损失函数影响的曲线
costs = np.squeeze(d['costs'])
plt.plot(costs)
plt.ylabel('cost')
plt.xlabel('iterations (per hundreds)')
plt.title("Learning rate =" + str(d["learning_rate"]))
plt.show()

测试自己的图片

my_image = "cat2.jpg"

fname = "C:\\Users\\董润泽\\Desktop\\cat_picture\\" + my_image
image = np.array(ndimage.imread(fname, flatten=False))
my_image = scipy.misc.imresize(image, size=(num_px,num_px)).reshape((1, num_px*num_px*3)).T
my_predicted_image,A = predict(d["w"], d["b"], my_image)

plt.imshow(image)
print("y = " + str(np.squeeze(my_predicted_image)) + ", your algorithm predicts a \"" + classes[int(np.squeeze(my_predicted_image)),].decode("utf-8") +  "\" picture.")
print(A[0][0])

y = 1.0, your algorithm predicts a "cat" picture.
1.0


C:\ProgramData\Anaconda3\lib\site-packages\ipykernel_launcher.py:4: DeprecationWarning: `imread` is deprecated!
`imread` is deprecated in SciPy 1.0.0.
Use ``matplotlib.pyplot.imread`` instead.
  after removing the cwd from sys.path.
C:\ProgramData\Anaconda3\lib\site-packages\ipykernel_launcher.py:5: DeprecationWarning: `imresize` is deprecated!
`imresize` is deprecated in SciPy 1.0.0, and will be removed in 1.3.0.
Use Pillow instead: ``numpy.array(Image.fromarray(arr).resize())``.
  """

---

转载请注明来源，欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论，也可以邮件至 2470290795@qq.com