use OrderedDict

Author: koki0702

ch07/simple_convnet.py

@@ -24,37 +24,42 @@ class SimpleConvNet:
         'sigmoid'または'xavier'を指定した場合は「Xavierの初期値」を設定
     """
     def __init__(self, input_dim=(1, 28, 28), 
-                 conv_param = {'filter_num':30, 'filter_size':5, 'pad':0, 'stride':1},
+                 conv_param={'filter_num':30, 'filter_size':5, 'pad':0, 'stride':1},
                  hidden_size=100, output_size=10, weight_init_std=0.01):
         filter_num = conv_param['filter_num']
         filter_size = conv_param['filter_size']
         filter_pad = conv_param['pad']
         filter_stride = conv_param['stride']
         input_size = input_dim[1]
         conv_output_size = (input_size - filter_size + 2*filter_pad) / filter_stride + 1
+        pool_output_size = int(filter_num * (conv_output_size/2) * (conv_output_size/2))
 
         # 重みの初期化
         self.params = {}
-        self.params['W1'] = weight_init_std * np.random.randn(filter_num, input_dim[0], filter_size, filter_size)
+        self.params['W1'] = weight_init_std * \
+                            np.random.randn(filter_num, input_dim[0], filter_size, filter_size)
         self.params['b1'] = np.zeros(filter_num)
-        self.params['W2'] = weight_init_std * np.random.randn(int(filter_num*(conv_output_size/2)*(conv_output_size/2)), hidden_size) 
+        self.params['W2'] = weight_init_std * \
+                            np.random.randn(pool_output_size, hidden_size)
         self.params['b2'] = np.zeros(hidden_size)
-        self.params['W3'] = weight_init_std * np.random.randn(hidden_size, output_size) 
+        self.params['W3'] = weight_init_std * \
+                            np.random.randn(hidden_size, output_size)
         self.params['b3'] = np.zeros(output_size)
 
         # レイヤの生成
-        self.layers = []
-        self.layers.append(Convolution(self.params['W1'], self.params['b1'], conv_param['stride'], conv_param['pad']))
-        self.layers.append(Relu())
-        self.layers.append(Pooling(pool_h=2, pool_w=2, stride=2))
-        self.layers.append(Affine(self.params['W2'], self.params['b2']))
-        self.layers.append(Relu())
-        self.layers.append(Affine(self.params['W3'], self.params['b3']))
-        
+        self.layers = OrderedDict()
+        self.layers['Conv1'] = Convolution(self.params['W1'], self.params['b1'],
+                                           conv_param['stride'], conv_param['pad'])
+        self.layers['Relu1'] = Relu()
+        self.layers['Pool1'] = Pooling(pool_h=2, pool_w=2, stride=2)
+        self.layers['Affine1'] = Affine(self.params['W2'], self.params['b2'])
+        self.layers['Relu2'] = Relu()
+        self.layers['Affine2'] = Affine(self.params['W3'], self.params['b3'])
+
         self.last_layer = SoftmaxWithLoss()
 
     def predict(self, x):
-        for layer in self.layers:
+        for layer in self.layers.values():
             x = layer.forward(x)
 
         return x
@@ -124,19 +129,16 @@ def gradient(self, x, t):
         dout = 1
         dout = self.last_layer.backward(dout)
 
-        tmp_layers = self.layers.copy()
-        tmp_layers.reverse()
-        for layer in tmp_layers:
+        layers = list(self.layers.values())
+        layers.reverse()
+        for layer in layers:
             dout = layer.backward(dout)
 
         # 設定
         grads = {}
-        grads['W1'] = self.layers[0].dW
-        grads['b1'] = self.layers[0].db
-        grads['W2'] = self.layers[3].dW
-        grads['b2'] = self.layers[3].db
-        grads['W3'] = self.layers[5].dW
-        grads['b3'] = self.layers[5].db
+        grads['W1'], grads['b1'] = self.layers['Conv1'].dW, self.layers['Conv1'].db
+        grads['W2'], grads['b2'] = self.layers['Affine1'].dW, self.layers['Affine1'].db
+        grads['W3'], grads['b3'] = self.layers['Affine2'].dW, self.layers['Affine2'].db
 
         return grads
 
@@ -153,6 +155,6 @@ def load_params(self, file_name="params.pkl"):
         for key, val in params.items():
             self.params[key] = val
 
-        for i, layer_idx in enumerate((0, 3, 5)):
-            self.layers[layer_idx].w = self.params['W' + str(i+1)]
-            self.layers[layer_idx].b = self.params['b' + str(i+1)]
+        for i, key in enumerate(['Conv1', 'Affine1', 'Affine2']):
+            self.layers[key].W = self.params['W' + str(i+1)]
+            self.layers[key].b = self.params['b' + str(i+1)]