Keras/Theano VGG16 AttributeError的：「模型」對象有沒有屬性「NDIM」

我試圖建立與最後一個時間分佈VGG16網絡連接到LSTM層，然後到一系列的Theano後端一個Keras模型密集的層。不過，我得到以下錯誤：Keras/Theano VGG16 AttributeError的：「模型」對象有沒有屬性「NDIM」

Traceback (most recent call last): 
    File "osr.py", line 341, in <module> 
    osr.generate_osr_model() 
    File "osr.py", line 145, in generate_osr_model 
    cnn_out = GlobalAveragePooling2D()(cnn_base) 
    File "/usr/local/lib/python2.7/dist-packages/keras/engine/topology.py", line 529, in __call__ 
    self.assert_input_compatibility(x) 
    File "/usr/local/lib/python2.7/dist-packages/keras/engine/topology.py", line 464, in assert_input_compatibility 
    if K.ndim(x) != spec.ndim: 
    File "/usr/local/lib/python2.7/dist-packages/keras/backend/theano_backend.py", line 142, in ndim 
    return x.ndim 
AttributeError: 'Model' object has no attribute 'ndim'

這裏是代碼的部分，我用它來構建模型：

 video = Input(shape=(self.frames_per_sequence, 
          3, 
          self.rows, 
          self.columns)) 
     cnn_base = VGG16(input_shape=(3, 
             self.rows, 
             self.columns), 
         weights="imagenet", 
         include_top=False) 
     cnn_out = GlobalAveragePooling2D()(cnn_base) 
     cnn = Model(input=cnn_base.input, output=cnn_out) 
     cnn.trainable = False 
     encoded_frames = TimeDistributed(cnn)(video) 
     encoded_vid = LSTM(256)(encoded_frames) 
     hidden_layer = Dense(output_dim=1024, activation="relu")(encoded_vid) 
     outputs = Dense(output_dim=class_count, activation="softmax")(hidden_layer) 
     osr = Model([video], outputs) 
     optimizer = Nadam(lr=0.002, 
          beta_1=0.9, 
          beta_2=0.999, 
          epsilon=1e-08, 
          schedule_decay=0.004) 
     osr.compile(loss="categorical_crossentropy", 
        optimizer=optimizer, 
        metrics=["categorical_accuracy"])

來源

2017-02-28 Robert Valencia

的解決方案是使用cnn_base.output作爲輸入到GlobalAveragePooling2D層：

cnn_out = GlobalAveragePooling2D()(cnn_base.output)

來源

2017-02-28 22:57:35

Keras/Theano VGG16 AttributeError的：「模型」對象有沒有屬性「NDIM」

回答

相關問題