一直試圖在Keras中建立神經網絡,但遇到了我的密集層和激活層之間存在形狀不匹配的問題。我錯過了明顯的東西嗎?使用Tensorflow後端。Keras:密集層和激活層之間的形狀不匹配
print(x_train.shape)
print(y_train.shape)
(1509, 476, 4)
(1509,)
然後我的模型如下:
###Setup Keras to create a bidirectional convolutional recurrent NN based on DanQ NN
###See https://github.com/uci-cbcl/DanQ
model = Sequential()
model.add(Conv1D(filters=320,
kernel_size=26,
padding="valid",
activation="relu",
strides=1,
input_shape=(476, 4)
))
model.add(MaxPooling1D(pool_size=13, strides=13))
model.add(Dropout(0.2))
model.add(keras.layers.wrappers.Bidirectional(LSTM(320, return_sequences=True, input_shape=(None, 320))))
model.add(Flatten())
model.add(Dense(input_dim=34*640, units=925))
model.add(Activation('relu'))
model.add(Dense(input_dim=925, units=919))
model.add(Activation('sigmoid'))
print('compiling model')
model.compile(loss='binary_crossentropy', optimizer='rmsprop', class_mode="binary")
print('running at most 60 epochs')
model.fit(x_train, y_train.T, batch_size=100, epochs=60, shuffle=True, verbose=2, validation_split=0.1)
tresults = model.evaluate(x_test, y_test, verbose=2)
print(tresults)
print(model.output_shape)
,但我得到了以下錯誤:
ValueError: Error when checking target: expected activation_48 to have shape (None, 919) but got array with shape (1509, 1)
的錯誤似乎從輸入使用始發到第二激活層一個S形激活。例如:
model.add(Dense(input_dim=925, units=919))
model.add(Activation('sigmoid'))
爲什麼會出現不匹配?
您不必在密集圖層中指定input_dim和單位。恩。最後密集層應該是'model.add(Dense(1,activation ='sigmoid'))' – DJK
謝謝!這似乎解決了不匹配問題,但彈出一個新的錯誤:'ValueError:無效的參數「class_mode」通過Tensorflow backend傳遞給K.function,源自'model.fit()'。對此有何想法? –
是的,你不需要在'model.compile()'中有'class_mode ='binary''參數' – DJK