The list of Numpy arrays that you are passing to your model is not the size the model expected on 3d arrays

Question

我正在学习使用keras的Sequential构建神经网络，但是运行此命令时出现错误：

X = [
    [[1,4,5],[7,8,15],[13,16,45],[19,32,135]],
    [[0,0,0],[1,4,9], [16,25,36],[49,64,82]],
    [[1,1,1], [2,2,2], [1,1,1], [2,2,2]]
]

y = [
 [[25,64,405], [31,128,1215]],
 [[100,121,144], [169,196, 225]],
 [[1,1,1],[2,2,2]]   
]
X= np.array(X)
Y= np.array(y)

model = keras.models.Sequential()
model.add(LSTM(16, activation='relu', input_shape=(4, 3)))
model.add(Dense(32, activation='relu'))
model.add(Dense(3))
model.compile(loss='mse', optimizer='adam')
# fit network
model.fit(X, y, epochs=10, batch_size=4)
return model

Error when checking model target: the list of Numpy arrays that you are passing to your model is not the size the model expected. Expected to see 1 array(s), but instead got the following list of 3 arrays:

我认为 Sequential 应该采用 3d 数组，我在其中指定了timesteps=4和features=3的数量，但看起来我的格式不正确。 谁能告诉我我做错了什么？

Answer 1

首先，您将 model 安装到y和到Y 。 其次，Y 应该是一个 (None, 3) 数组，因为您的网络输出 3 个值，但 Y 是 (None, 2, 3)。

从纠正这些错误开始，然后看看是否有问题