一、網絡層layer的常見操作
通常機器學習模型可以表示為簡單網絡層的堆疊與組合，而tensorflow就提供了常見的網絡層，為我們編寫神經網絡程序提供了便利。
TensorFlow2推薦使用tf.keras來構建網絡層，tf.keras來自原生keras，用其來構建網絡具有更好的可讀性和易用性。
如，我們要構造一個簡單的全連接網絡，只需要指定網絡的神經元個數
layer = tf.keras.layers.Dense(100)
# 也可以添加輸入維度限制
layer = tf.keras.layers.Dense(100, input_shape=(None, 20))
可以在文檔中查看預先存在的圖層的完整列表。 它包括Dense，Conv2D，LSTM，BatchNormalization，Dropout等等。
每個層都可以當作一個函數，然后以輸入的數據作為函數的輸入
layer(tf.ones([6, 6]))
同時我們也可以得到網絡的變量、權重矩陣、偏置等
print(layer.variables) # 包含了權重和偏置
[
array([[-0.18237606, 0.16415142, 0.20687856, 0.23396944, 0.09779547,
-0.14794639, -0.10231382, -0.22263053, -0.0950674 , 0.18697281,
...
-0.10450925, -0.12111329, -0.2259491 , 0.12304659, -0.04047236]],
dtype=float32)>,
array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
dtype=float32)>]
print(layer.kernel, layer.bias) # 也可以分別取出權重和偏置
array([[-0.18237606, 0.16415142, 0.20687856, 0.23396944, 0.09779547,
...
-0.10450925, -0.12111329, -0.2259491 , 0.12304659, -0.04047236]],
dtype=float32)>
array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
dtype=float32)>
二、實現自定義網絡層
實現自己的層的最佳方法是擴展tf.keras.Layer類並實現：
_init_()函數，你可以在其中執行所有與輸入無關的初始化
build()函數，可以獲得輸入張量的形狀，並可以進行其余的初始化
call()函數，構建網絡結構，進行前向傳播
實際上，你不必等到調用build()來創建網絡結構，您也可以在_init_()中創建它們。 但是，在build()中創建它們的優點是它可以根據圖層將要操作的輸入的形狀啟用后期的網絡構建。 另一方面，在__init__中創建變量意味着需要明確指定創建變量所需的形狀。
class MyDense(tf.keras.layers.Layer):
def __init__(self, n_outputs):
super(MyDense, self).__init__()
self.n_outputs = n_outputs
def build(self, input_shape):
self.kernel = self.add_variable('kernel',
shape=[int(input_shape[-1]),
self.n_outputs])
def call(self, input):
return tf.matmul(input, self.kernel)
layer = MyDense(10)
print(layer(tf.ones([6, 5])))
print(layer.trainable_variables)
tf.Tensor(
[[ 1.0200843 -0.42590106 -0.92992705 0.46160045 0.7518406 0.32543844
0.34020287 0.08215448 0.22044104 -0.5337319 ]
[ 1.0200843 -0.42590106 -0.92992705 0.46160045 0.7518406 0.32543844
0.34020287 0.08215448 0.22044104 -0.5337319 ]
[ 1.0200843 -0.42590106 -0.92992705 0.46160045 0.7518406 0.32543844
0.34020287 0.08215448 0.22044104 -0.5337319 ]
[ 1.0200843 -0.42590106 -0.92992705 0.46160045 0.7518406 0.32543844
0.34020287 0.08215448 0.22044104 -0.5337319 ]
[ 1.0200843 -0.42590106 -0.92992705 0.46160045 0.7518406 0.32543844
0.34020287 0.08215448 0.22044104 -0.5337319 ]
[ 1.0200843 -0.42590106 -0.92992705 0.46160045 0.7518406 0.32543844
0.34020287 0.08215448 0.22044104 -0.5337319 ]], shape=(6, 10), dtype=float32)
[
array([[ 0.54810244, 0.042225 , 0.25634396, 0.1677258 , -0.0361526 ,
0.32831818, 0.17709464, 0.46625894, 0.29662275, -0.32920587],
[ 0.30925363, -0.426274 , -0.49862564, 0.3068235 , 0.29526353,
0.50076336, 0.17321467, 0.21151704, -0.26317668, -0.2006711 ],
[ 0.10354012, -0.3258371 , -0.12274069, -0.33250242, 0.46343058,
-0.45535576, 0.5332853 , -0.37351888, -0.00410944, 0.16418225],
[-0.4515978 , 0.04706419, -0.42583126, -0.19347438, 0.54246336,
0.57910997, 0.01877069, 0.01255274, -0.14176458, -0.6309683 ],
[ 0.5107859 , 0.23692083, -0.13907343, 0.51302797, -0.5131643 ,
-0.6273973 , -0.56216246, -0.23465535, 0.332869 , 0.4629311 ]],
dtype=float32)>]
三、網絡層組合
機器學習模型中有很多是通過疊加不同的結構層組合而成的，如resnet的每個殘差塊就是“卷積+批標准化+殘差連接”的組合。
在tensorflow2中要創建一個包含多個網絡層的的結構，一般繼承與tf.keras.Model類。
# 殘差塊
class ResnetBlock(tf.keras.Model):
def __init__(self, kernel_size, filters):
super(ResnetBlock, self).__init__(name='resnet_block')
# 每個子層卷積核數
filter1, filter2, filter3 = filters
# 三個子層，每層1個卷積加一個批正則化
# 第一個子層， 1*1的卷積
self.conv1 = tf.keras.layers.Conv2D(filter1, (1,1))
self.bn1 = tf.keras.layers.BatchNormalization()
# 第二個子層， 使用特點的kernel_size
self.conv2 = tf.keras.layers.Conv2D(filter2, kernel_size, padding='same')
self.bn2 = tf.keras.layers.BatchNormalization()
# 第三個子層，1*1卷積
self.conv3 = tf.keras.layers.Conv2D(filter3, (1,1))
self.bn3 = tf.keras.layers.BatchNormalization()
def call(self, inputs, training=False):
# 堆疊每個子層
x = self.conv1(inputs)
x = self.bn1(x, training=training)
x = self.conv2(x)
x = self.bn2(x, training=training)
x = self.conv3(x)
x = self.bn3(x, training=training)
# 殘差連接
x += inputs
outputs = tf.nn.relu(x)
return outputs
resnetBlock = ResnetBlock(2, [6,4,9])
# 數據測試
print(resnetBlock(tf.ones([1,3,9,9])))
# 查看網絡中的變量名
print([x.name for x in resnetBlock.trainable_variables])
tf.Tensor(無錫婦科醫院排行 http://www.0510bhyy.com/
[[[[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.83203167 0.9436392 1.0989372 1.2588525 0.8683256 1.1279813
0.7571581 0.47963202 0.88908756]]
[[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.79764616 1.0550306 0.9386751 1.1079601 0.9402881 0.99479383
0.9072118 0.5618475 0.9134829 ]
[0.83203167 0.9436392 1.0989372 1.2588525 0.8683256 1.1279813
0.7571581 0.47963202 0.88908756]]
[[1.0775117 1.1620466 0.72680396 1.0019443 1.2767658 1.1365149
1.1792164 1.0868194 1.0623009 ]
[1.0775117 1.1620466 0.72680396 1.0019443 1.2767658 1.1365149
1.1792164 1.0868194 1.0623009 ]
[1.0775117 1.1620466 0.72680396 1.0019443 1.2767658 1.1365149
1.1792164 1.0868194 1.0623009 ]
[1.0775117 1.1620466 0.72680396 1.0019443 1.2767658 1.1365149
1.1792164 1.0868194 1.0623009 ]
[1.0775117 1.1620466 0.72680396 1.0019443 1.2767658 1.1365149
1.1792164 1.0868194 1.0623009 ]
[1.0775117 1.1620466 0.72680396 1.0019443 1.2767658 1.1365149
1.1792164 1.0868194 1.0623009 ]
[1.0775117 1.1620466 0.7268039 1.0019443 1.2767658 1.1365149
1.1792164 1.0868194 1.0623009 ]
[1.0775117 1.1620466 0.7268039 1.0019443 1.2767658 1.1365149
1.1792164 1.0868194 1.0623009 ]
[0.87889266 0.9541194 0.8929231 0.96703756 1.0905087 1.0646607
0.9235744 0.9829142 1.1302696 ]]]], shape=(1, 3, 9, 9), dtype=float32)
['resnet_block/conv2d_12/kernel:0', 'resnet_block/conv2d_12/bias:0', 'resnet_block/batch_normalization_v2_12/gamma:0', 'resnet_block/batch_normalization_v2_12/beta:0', 'resnet_block/conv2d_13/kernel:0', 'resnet_block/conv2d_13/bias:0', 'resnet_block/batch_normalization_v2_13/gamma:0', 'resnet_block/batch_normalization_v2_13/beta:0', 'resnet_block/conv2d_14/kernel:0', 'resnet_block/conv2d_14/bias:0', 'resnet_block/batch_normalization_v2_14/gamma:0', 'resnet_block/batch_normalization_v2_14/beta:0']
如果模型是線性的，可以直接用tf.keras.Sequential來構造。
seq_model = tf.keras.Sequential(
[
tf.keras.layers.Conv2D(1, 1, input_shape=(None, None, 3)),
tf.keras.layers.BatchNormalization(),
tf.keras.layers.Conv2D(2, 1, padding='same'),
tf.keras.layers.BatchNormalization(),
tf.keras.layers.Conv2D(3, 1),
tf.keras.layers.BatchNormalization(),
])
seq_model(tf.ones([1,2,3,3]))
array([[[[-0.36850607, -0.60731524, 1.2792252 ],
[-0.36850607, -0.60731524, 1.2792252 ],
[-0.36850607, -0.60731524, 1.2792252 ]],
[[-0.36850607, -0.60731524, 1.2792252 ],
[-0.36850607, -0.60731524, 1.2792252 ],
[-0.36850607, -0.60731524, 1.2792252 ]]]], dtype=float32)>