keras：ValueError：檢查模型目標時出錯：預期activation_1具有形狀（無，60），但數組的形狀為（10，100）

Question

我正在嘗試將RocAlphaGo移植到“亞馬遜游戲”中，並且在嘗試實施監督策略培訓師時會遇到問題。

from keras.models import Sequential, Model
from keras.layers.core import Activation, Flatten
from keras.layers import convolutional

defaults = {
            "board": 10,
            "filters_per_layer": 128,
            "layers": 12,
            "filter_width_1": 5
}
# copy defaults, but override with anything in kwargs
params = defaults
network = Sequential()
# create first layer
network.add(convolutional.Convolution2D(
      input_shape=(6, 10, 10),
      nb_filter=128,
      nb_row=5,
      nb_col=5,
      init='uniform',
      activation='relu',
      border_mode='same'))

# create all other layers
for i in range(2, 13):
     # use filter_width_K if it is there, otherwise use 3
     filter_key = "filter_width_%d" % i
     filter_width = params.get(filter_key, 3)

     # use filters_per_layer_K if it is there, otherwise use default value
     filter_count_key = "filters_per_layer_%d" % i
     filter_nb = params.get(filter_count_key, 128)

     network.add(convolutional.Convolution2D(
         nb_filter=filter_nb,
         nb_row=filter_width,
         nb_col=filter_width,
         init='uniform',
         activation='relu',
         border_mode='same'))

# the last layer maps each <filters_per_layer> feature to a number
network.add(convolutional.Convolution2D(
    nb_filter=1,
    nb_row=1,
    nb_col=1,
    init='uniform',
    border_mode='same'))
# reshape output to be board x board
network.add(Flatten())
# softmax makes it into a probability distribution
network.add(Activation('softmax'))

• keras 1.2.0
• python 2.7

給出以下異常：

ValueError：檢查模型目標時出錯：預期activation_1的形狀為（無，60），但數組的形狀為（10，100）

訓練數據集是一個（10、6、10、10）數組，每層10 x 6層，每層是10x10數組（棋盤），為什么需要模型（無，60）？ 如果chagne input_shape=(6, 10, 10)到input_shape=(10, 10, 10) ，將得到：

ValueError：檢查模型輸入時出錯：預期convolution2d_input_1具有形狀（None，10，10，10）但具有形狀（10，6，10，10）的數組

所有代碼都在這里

Answer 1

正如Matias在評論中所說，如果您添加

network.summary() 

您可能會注意到，將卷積應用於輸入數據的前兩個維度（即（6,10,10））。 您的功能在第一個維度。 默認情況下，當您使用tensorflow時，Keras將假定要素將要放置的尺寸是第三個尺寸，而不是第一個尺寸。 出於這個原因，當您在（6,10,10）數組上應用128個過濾器時，輸出將是（6,10,128），如果我理解的話，這不是您想要的。

因此，在最后一個卷積層的輸出處，您將獲得一個（6,10,1）數組，該數組將展平為（，60）而不是您期望的（100）。

有兩種修復網絡的方法。 您可以將輸入數據更改為（10,10,6）格式。 或者，您可以使用Convolution2D()層的data_format="channels_first"參數。

這是第二個選項的代碼：

from keras.models import Sequential, Model
from keras.layers.core import Activation, Flatten
from keras.layers import convolutional

defaults = {
            "board": 10,
            "filters_per_layer": 128,
            "layers": 12,
            "filter_width_1": 5
}
# copy defaults, but override with anything in kwargs
params = defaults
network = Sequential()
# create first layer
network.add(convolutional.Convolution2D(
      input_shape=(6, 10, 10),
      nb_filter=128,
      nb_row=5,
      nb_col=5,
      init='uniform',
      activation='relu',
      border_mode='same',
      data_format='channels_first'
      ))

# create all other layers
for i in range(2, 13):
     # use filter_width_K if it is there, otherwise use 3
     filter_key = "filter_width_%d" % i
     filter_width = params.get(filter_key, 3)

     # use filters_per_layer_K if it is there, otherwise use default value
     filter_count_key = "filters_per_layer_%d" % i
     filter_nb = params.get(filter_count_key, 128)

     network.add(convolutional.Convolution2D(
         nb_filter=filter_nb,
         nb_row=filter_width,
         nb_col=filter_width,
         init='uniform',
         activation='relu',
         border_mode='same',
         data_format='channels_first'))

# the last layer maps each <filters_per_layer> feature to a number
network.add(convolutional.Convolution2D(
    nb_filter=1,
    nb_row=1,
    nb_col=1,
    init='uniform',
    border_mode='same',
    data_format='channels_first'))
# reshape output to be board x board
network.add(Flatten())
# softmax makes it into a probability distribution
network.add(Activation('softmax'))
# display your network summary
network.summary()

編輯

考慮到您的keras版本，您應該使用參數“ dim_ordering”並將其設置為“ th”。

我在keras文檔中找到了此信息

from keras.models import Sequential, Model
from keras.layers.core import Activation, Flatten
from keras.layers import convolutional

defaults = {
            "board": 10,
            "filters_per_layer": 128,
            "layers": 12,
            "filter_width_1": 5
}
# copy defaults, but override with anything in kwargs
params = defaults
network = Sequential()
# create first layer
network.add(convolutional.Convolution2D(
      input_shape=(6, 10, 10),
      nb_filter=128,
      nb_row=5,
      nb_col=5,
      init='uniform',
      activation='relu',
      border_mode='same',
      dim_ordering='th'
      ))

# create all other layers
for i in range(2, 13):
     # use filter_width_K if it is there, otherwise use 3
     filter_key = "filter_width_%d" % i
     filter_width = params.get(filter_key, 3)

     # use filters_per_layer_K if it is there, otherwise use default value
     filter_count_key = "filters_per_layer_%d" % i
     filter_nb = params.get(filter_count_key, 128)

     network.add(convolutional.Convolution2D(
         nb_filter=filter_nb,
         nb_row=filter_width,
         nb_col=filter_width,
         init='uniform',
         activation='relu',
         border_mode='same',
         dim_ordering='th'))

# the last layer maps each <filters_per_layer> feature to a number
network.add(convolutional.Convolution2D(
    nb_filter=1,
    nb_row=1,
    nb_col=1,
    init='uniform',
    border_mode='same',
    dim_ordering='th'))
# reshape output to be board x board
network.add(Flatten())
# softmax makes it into a probability distribution
network.add(Activation('softmax'))
# display your network summary
network.summary()