如何将以列表形式出现的一组值传递到神经网络并预测其输出

Question

我有一个列表[0 0 0 0 1 0 1 1 0 1 0 0 0 0 1]，我想把它传递给一个神经网络来预测它的输出..怎么做

我试着把它转换成一个数组，但是没有用，因为它的维数不合适

classifier = Sequential()

classifier.add(Dense(output_dim = 64, init = 'uniform', activation= 'relu', input_dim = 15))

#adding second hidden layer
classifier.add(Dense(output_dim = 64, init = 'uniform', activation= 'relu'))

#adding last layer
classifier.add(Dense(output_dim = 28, init = 'uniform', activation= 'softmax'))

#compiling ANN
classifier.compile(optimizer = 'adam', loss = 'categorical_crossentropy', metrics = ['accuracy'])

#fitting ann to training set
classifier.fit(xtrain, ytrain, batch_size =10, nb_epoch = 10)


testarray = [0, 0, 0, 0, 1, 0, 1, 1, 0, 1, 0, 0, 0, 0, 1]
classifier.predict(testarray)

我希望它预测一个数字，但它显示了错误消息

Error when checking input: expected dense_13_input to have shape (15,) but got array with shape (1,)