Coursera-吴恩达-深度学习-神经网络和深度学习-week4-编程1

本文章内容：

Coursera吴恩达深度学习课程，第一课神经网络和深度学习Neural Networks and Deep Learning，

第四周：深层神经网络(Deep Neural Networks)

编程作业: Building your deep neural network: Step by Step

错题本

 

我的;

parameters['W' + str(l)] = np.random.randn(L[l], L[l-1]) * 0.01

        parameters['b' + str(l)] = np.zeros((L[l]),1)

正确：

parameters['W' + str(l)] = np.random.randn(layer_dims[l], layer_dims[l-1])*0.01

        parameters['b' + str(l)] = np.zeros( (layer_dims[l], 1) )

反思：

取数组的元素，不是循环长度的。

 

我的：

# Implement [LINEAR -> RELU]*(L-1). Add "cache" to the "caches" list.

    for l in range(1, L):

        A_prev = A

        ### START CODE HERE ### (≈ 2 lines of code)

        A, cache = linear_activation_forward(A_prev, W, b, activation = "relu")

        None

        ### END CODE HERE ###

    

    # Implement LINEAR -> SIGMOID. Add "cache" to the "caches" list.

    ### START CODE HERE ### (≈ 2 lines of code)

    AL, cache = linear_activation_forward(A_prev, W, b, activation = "sigmoid")

    None

正确：

# Implement [LINEAR -> RELU]*(L-1). Add "cache" to the "caches" list.

    for l in