Question

我想创建一个keras模型，该模型执行以下操作：该模型作为输入x_t并预测y_t1 现在使用 y_t1（模型输入）预测y_t2 ..... y_t3（模型输入）可预测y_t4，依此类推。我是keras功能api的新手，不确定下面的代码是否满足我的要求。

 def model():
    input = Input((self.data.shape[1], self.data.shape[2]))
    print((input))
    print('shape ({}, {})'.format(self.data.shape[1], self.data.shape[2]))
    x = LSTM(128, activation='relu', return_sequences=True)(input)
    x = Dropout(0.2)(x)
    x = LSTM(128, activation='relu')(x)
    x = Dropout(0.2)(x)
    x = Dense(32, activation='relu')(x)
    x = Dropout(0.2)(x)
    x = Dense((21), activation='relu')(x)
    self.model = Model(input, x)
    for t in range(self.time_step):
        x = Reshape(target_shape=(1, 21))(x)
        x = self.model(x) 
    self.model = Model(input, x)

keras的n步预测

0 个答案: