DL之DNN：自定义2层神经网络TwoLayerNet模型(封装为层级结构)利用MNIST数据集进行训练、GC对比

输出结果

设计思路

核心代码

class TwoLayerNet:

   def __init__(self, input_size, hidden_size, output_size, weight_init_std = 0.01):

       self.params = {}

       self.params['W1'] = weight_init_std * np.random.randn(input_size, hidden_size)

       self.params['b1'] = np.zeros(hidden_size)

       self.params['W2'] = weight_init_std * np.random.randn(hidden_size, output_size)

       self.params['b2'] = np.zeros(output_size)

       self.layers = OrderedDict()

       self.layers['Affine1'] = Affine(self.params['W1'], self.params['b1'])

       self.layers['Relu1'] = Relu()

       self.layers['Affine2'] = Affine(self.params['W2'], self.params['b2'])

       self.lastLayer = SoftmaxWithLoss()

   def predict(self, x):

       for layer in self.layers.values():

           x = layer.forward(x)

       return x

   # x:输入数据, t:监督数据

   def loss(self, x, t):

       y = self.predict(x)

       return self.lastLayer.forward(y, t)

   def accuracy(self, x, t):

       y = self.predict(x)

       y = np.argmax(y, axis=1)

       if t.ndim != 1 : t = np.argmax(t, axis=1)

       accuracy = np.sum(y == t) / float(x.shape[0])

       return accuracy

   def gradient(self, x, t):

       self.loss(x, t)

       dout = 1

       dout = self.lastLayer.backward(dout)

       layers = list(self.layers.values())

       layers.reverse()

       for layer in layers:

           dout = layer.backward(dout)

       grads = {}

       grads['W1'], grads['b1'] = self.layers['Affine1'].dW, self.layers['Affine1'].db

       grads['W2'], grads['b2'] = self.layers['Affine2'].dW, self.layers['Affine2'].db

       return grads

network_batch = TwoLayerNet(input_size=784, hidden_size=50, output_size=10)

grad_numerical = network_batch.numerical_gradient_api(x_batch, t_batch)  

grad_backprop = network_batch.gradient(x_batch, t_batch)