Блог

Это простая модель:

 import numpy as np
import copy

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

Я инициализирую веса и смещение как нули
Эта функция создает вектор нулей формы (dim, 1) для w и инициализирует b до 0.

Аргумент: dim — размер вектора w, который мы хотим (или количество параметров в данном случае)

Возвращает: w — инициализированный вектор формы (dim, 1) b — инициализированный скаляр (соответствует смещению) типа float

 def initialize_with_zeros(dim):
        w = np.zeros((dim,1))
        b = 0.
        return w,b
 

Эта функция предназначена для прямого и обратного прохода

 def propagate(w, b, X, Y):

    m = X.shape[1]
    # epsilon = 1e-5
    A = sigmoid(np.dot(w.T,X)   b)
    cost = (-1/m) * (np.dot(Y,np.log(A).T)   np.dot((1-Y),np.log(1-A).T))

    dw = 1 / m *(np.dot(X,(A - Y ).T))
    db = 1/m * (np.sum(A-Y))
    
    cost = np.squeeze(np.array(cost))
    
    grads = {"dw": dw,
             "db": db}
    
    return grads, cost
 

Эта функция предназначена для обновления параметров

 def optimize(w,b,X,Y,num_iterations=100,learning_rate=0.009,print_cost=False):
    w = copy.deepcopy(w)
    b = copy.deepcopy(b)

    costs = []
    for i in range(num_iterations):

        #gradient and cost calculations
        grads, cost = propagate(w,b,X,Y)

        #retrieve derivatives from grads
        dw = grads["dw"]
        db = grads["db"]

        #update 
        w = w - learning_rate * dw
        b = b - learning_rate * db

        #record cost every 100th iterations
        if i0==0:
            costs.append(cost)

            if print_cost:
                print("cost after %i:%f" %(i,cost))
    
    params = {"w": w,
              "b": b}
    
    grads = {"dw": dw,
             "db": db}
    
    return params, grads, costs
 

This function is for predicting the result

 def predict(w,b,X):
    
    m = X.shape[1]
    Y_prediction = np.zeros((1,m))
    w = w.reshape(X.shape[0],1)

    A = sigmoid(np.dot(w.T,X) b)
    
    for i in range(A.shape[1]):
        if A[0][i] > 0.5:
            Y_prediction[0][i] = 1.0
        else:
            Y_prediction[0][i] = 0.0

    return Y_prediction
 

Logistic regression model :

 def logisticregressionmodel(X_train, Y_train, X_test, Y_test, num_iterations=2000, learning_rate=0.5, print_cost=False):

    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 = predict(w, b, X_test)
    Y_prediction_train = predict(w, b, X_train)

    # Print train/test Errors
    if print_cost:
        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


#result
 cost after 0:0.693147
cost after 100:nan
cost after 200:nan
cost after 300:nan
cost after 400:nan
cost after 500:nan
cost after 600:nan
cost after 700:nan
cost after 800:nan