吴恩达《深度学习》-课后测验-第一门课 (Neural Networks and Deep Learning)-Week 4 - Key concepts on Deep Neural Networks（第四周 测验 – 深层神经网络）
Week 4 Quiz - Key concepts on Deep Neural Networks（第四周 测验 – 深层神经网络）
\1. What is the “cache” used for in our implementation of forward propagation and backward propagation?(在实现前向传播和反向传播中使用的“cache”是什么？)
【 】It is used to cache the intermediate values of the cost function during training.(用于在训练期间缓存成本函数的中间值。)
【 】We use it to pass variables computed during forward propagation to the corresponding backward propagation step. It contains useful values for backward propagation to compute derivatives.(我们用它传递前向传播中计算的变量到相应的反向传播步骤，它包含用于计算导数的反向传播的有用值。)
【 】It is used to keep track of the hyperparameters that we are searching over, to speed up computation.(它用于跟踪我们正在搜索的超参数，以加速计算。)
【 】We use it to pass variables computed during backward propagation to the corresponding forward propagation step. It contains useful values for forward propagation to compute activations.(我们使用它将向后传播计算的变量传递给相应的正向传播步骤，它包含用于计算计算激活的正向传播的有用值。)
答案
【★】We use it to pass variables computed during forward propagation to the corresponding backward propagation step. It contains useful values for backward propagation to compute derivatives.(我们用它传递前向传播中计算的变量到相应的反向传播步骤，它包含用于计算导数的反向传播的有用值。)
Note: the “cache” records values from the forward propagation units and sends it to the backward propagation units because it is needed to compute the chain rule derivatives.(请注意：“cache”记录来自正向传播单元的值并将其发送到反向传播单元，因为需要链式计算导数。)
\2. Among the following, which ones are “hyperparameters”? (Check all that apply.)(以下哪些是“超参数”？)
【 】size of the hidden layers \(𝑛^{[𝑙]}\) (隐藏层的大小\(𝑛^{[𝑙]}\))
【 】learning rate α(学习率 α)
【 】number of iterations(迭代次数)
【 】number of layers 𝐿 in the neural network(神经网络中的层数𝐿)
答案
全对
Note: You can check this Quora post or this blog post.(请注意：你可以查看 Quora 的这篇文章 或者 这篇博客.)
\3. Which of the following statements is true?(下列哪个说法是正确的？)
【 】The deeper layers of a neural network are typically computing more complex features of the input than the earlier layers. (神经网络的更深层通常比前面的层计算更复杂的输入特征。) 【 】 The earlier layers of a neural network are typically computing more complex features of the input than the deeper layers.(神经网络的前面的层通常比更深层计算更复杂的输入特征。)
答案
【★】The deeper layers of a neural network are typically computing more complex features of the input than the earlier layers. (神经网络的更深层通常比前面的层计算更复杂的输入特征。)
Note: You can check the lecture videos. I think Andrew used a CNN example to explain this.(注意：您 可以查看视频，我想用吴恩达的用美国有线电视新闻网的例子来解释这个。)
\4. Vectorization allows you to compute forward propagation in an 𝑳-layer neural network without an explicit for-loop (or any other explicit iterative loop) over the layers l=1, 2, …,L. True/False?(向量化允许您在𝑳层神经网络中计算前向传播，而不需要在层(l = 1,2，…，L)上显式的使用 for-loop（或任何其他显式迭代循环），正确吗？)
【 】 True(正确) 【 】 False(错误)
答案
【★】 False(错误)
Note: We cannot avoid the for-loop iteration over the computations among layers.(请注意：在层间 计算中，我们不能避免 for 循环迭代。)
\5. Assume we store the values for \(𝑛^{[𝑙]}\) in an array called layers, as follows: layer_dims = \([𝑛_𝑥, 4,3,2,1]\). So layer 1 has four hidden units, layer 2 has 3 hidden units and so on. Which of the following for-loops will allow you to initialize the parameters for the model?(假设我们将 \(𝑛^{[𝑙]}\)的值存储在名为 layers 的数组中，如下所示：layer_dims =\([𝑛_𝑥, 4,3,2,1]\)。 因此，第 1 层有四个隐藏单元，第 2 层有三个隐藏单元，依此类推。您可以使用哪个 for 循环初始化模型参 数？)
答案
for(i in range(1, len(layer_dims))):
parameter[‘W’ + str(i)] = np.random.randn(layers[i], layers[i- 1])) * 0.01
parameter[‘b’ + str(i)] = np.random.randn(layers[i], 1) * 0.01
\6. Consider the following neural network.(下面关于神经网络的说法正确的是：)
【 】The number of layers𝐿 is 4. The number of hidden layers is 3.(层数𝐿为 4，隐藏层数为 3)
答案
显然
Note: The input layer (𝐿 [0] ) does not count.(注意：输入层（𝐿 [0]）不计数。)
As seen in lecture, the number of layers is counted as the number of hidden layers + 1. The input and
output layers are not counted as hidden layers.(正如视频中所看到的那样，层数被计为隐藏层数
+1。输入层和输出层不计为隐藏层。)
\7. During forward propagation, in the forward function for a layer 𝒍 you need to know what is the activation function in a layer (Sigmoid, tanh, ReLU, etc.). During backpropagation, the corresponding backward function also needs to know what is the activation function for layer 𝒍, since the gradient depends on it. True/False?(在前向传播期间，在层𝒍的前向传播函数中， 您需要知道层𝒍中的激活函数（Sigmoid，tanh，ReLU 等）是什么， 在反向传播期间，相应 的反向传播函数也需要知道第𝒍层的激活函数是什么，因为梯度是根据它来计算的，正确吗？)
【 】 True(正确) 【 】False(错误)
答案
True(正确)
Note: During backpropagation you need to know which activation was used in the forward propagation to be able to compute the correct derivative.(注：在反向传播期间，您需要知道正向传播中使用哪种激活函数才能计算正确的导数。)
8.There are certain functions with the following properties:(有一些函数具有以下属性：)
(i) To compute the function using a shallow network circuit, you will need a large network (where we measure size by the number of logic gates in the network), but (ii) To compute it using a deep network circuit, you need only an exponentially smaller network. True/False?((i) 使用浅网络电路计算函数时，需要一个大网络（我们通过网络中的逻辑门数量来度量大小）， 但是（ii）使用深网络电路来计算它，只需要一个指数较小的网络。真/假？)
【 】True(正确) 【 】False(错误)
答案
True
Note: See lectures, exactly same idea was explained.(参见视频，完全相同的题。)
\9. Consider the following 2 hidden layer neural network: Which of the following statements are True? (Check all that apply).((在 2 层隐层神经网络中，下列哪个说法是正确的？只列出了正 确选项))
【 】\(𝑊^{[1]}\) will have shape (4, 4)(\(𝑊^{[1]}\)的维度为 (4, 4))
【 】\(𝑏^{[1]}\) will have shape (4, 1)(\(𝑏^{[1]}\)的维度为 (4, 1))
【 】\(𝑊^{[2]}\) will have shape (3, 4)(\(𝑊^{[2]}\)的维度为 (3, 4))
【 】\(𝑏^{[2]}\) will have shape (3, 1)(\(𝑏^{[2]}\)的维度为 (3, 1))
【 】\(𝑏^{[3]}\) will have shape (1, 1)(\(𝑏^{[3]}\)的维度为 (1, 1))
【 】\(𝑊^{[3]}\) will have shape (1, 3)(\(𝑊^{[3]}\)的维度为 (1, 3))
答案
Note: See [this image] for general formulas.(注：请参阅图片。)
\10. Whereas the previous question used a specific network, in the general case what is the dimension of \(𝑾^{[1]}\) , the weight matrix associated with layer 𝒍?(前面的问题使用了一个特定的 网络，与层 ll 有关的权重矩阵在一般情况下，\(𝑾^{[1]}\) 的维数是多少)
【 】\(𝑊^{[1]}\) has shape (\(𝑛^{[𝑙]} ,𝑛^{[𝑙−1]} )(𝑊^{[1]}\)的维度是$ (𝑛^{[𝑙]} ,𝑛^{[𝑙−1]} )$
答案
Note: See [this image] for general formulas.(注：请参阅图片。)
