如何向 CNN+LSTM 网络添加额外数据答案

【问题标题】：How to add additional data to CNN+LSTM network如何向 CNN+LSTM 网络添加额外数据
【发布时间】：2020-09-17 13:00:15
【问题描述】：

我有以下网络（预训练的 CNN + LSTM 对视频进行分类）：

 frames, channels, rows, columns = 5,3,224,224

  video = Input(shape=(frames,
                      rows,
                      columns,
                      channels))
  
  cnn_base = VGG16(input_shape=(rows,
                                columns,
                                channels),
                  weights="imagenet",
                  include_top=True) #<=== include_top=True
  cnn_base.trainable = False

  cnn = Model(cnn_base.input, cnn_base.layers[-3].output, name="VGG_fm") # -3 is the 4096 layer
  encoded_frames = TimeDistributed(cnn , name = "encoded_frames")(video)
  encoded_sequence = LSTM(256, name = "encoded_seqeunce")(encoded_frames)
  hidden_layer = Dense(1024, activation="relu" , name = "hidden_layer")(encoded_sequence)
  outputs = Dense(10, activation="softmax")(hidden_layer)

  model = Model(video, outputs)

看起来像这样：

现在，我想将视频的 784 个特征的一维向量添加到最后一层。我尝试将最后两行替换为：

  encoding_input = keras.Input(shape=(784,), name="Encoding", dtype='float') 
  sentence_features = layers.Dense(units = 60, name = 'sentence_features')(encoding_input)
  x = layers.concatenate([sentence_features, hidden_layer])
  outputs = Dense(10, activation="softmax")(x)

但得到了错误：

ValueError: Graph disconnected: cannot obtain value for tensor Tensor("Sentence-Input-Encoding_3:0", shape=(None, 784), dtype=float32) at layer "sentence_features". The following previous layers were accessed without issue: ['encoded_frames', 'encoded_seqeunce']

任何建议：

【问题讨论】：

标签： python keras deep-learning keras-layer

【解决方案1】：

您的网络现在有两个输入...不要忘记将两者都传递给您的模型

model = Model([video,encoding_input], outputs)

完整示例

frames, channels, rows, columns = 5,3,224,224

video = Input(shape=(frames,
                  rows,
                  columns,
                  channels))

cnn_base = VGG16(input_shape=(rows,
                            columns,
                            channels),
              weights="imagenet",
              include_top=True)
cnn_base.trainable = False

cnn = Model(cnn_base.input, cnn_base.layers[-3].output, name="VGG_fm")
encoded_frames = TimeDistributed(cnn , name = "encoded_frames")(video)
encoded_sequence = LSTM(256, name = "encoded_seqeunce")(encoded_frames)
hidden_layer = Dense(1024, activation="relu" , name = "hidden_layer")(encoded_sequence)

encoding_input = Input(shape=(784,), name="Encoding", dtype='float') 
sentence_features = Dense(units = 60, name = 'sentence_features')(encoding_input)
x = concatenate([sentence_features, hidden_layer])
outputs = Dense(10, activation="softmax")(x)

model = Model([video,encoding_input], outputs) #<=== double input
model.summary()

【讨论】：