层顺序的输入 0 与层不兼容：

Question

I'm trying to predict a basic stock prices model. Here is my code

data = pd.read_csv("total_cases.csv")
x = data["date"]
world_cases = data["Turkey"].fillna(0)
time = np.arange(len(world_cases), dtype="float32")

split_time = 200
x_train = time[:split_time]
x_valid = time[split_time:]

y_train = world_cases[:split_time]
y_valid = world_cases[split_time:]
window_size = 20
batch_size = 32
shuffle_buffer_size=1000
train_data = tf.data.Dataset.from_tensor_slices((x_train, y_train))
valid_data = tf.data.Dataset.from_tensor_slices((x_valid, y_valid))
model = Sequential()
model.add(LSTM(16, return_sequences=True))

model.add(LSTM(16))

model.add(Dense(16, activation='relu'))
model.compile(optimizer='adam', loss='mae', metrics=['mae'])

r = model.fit(train_data, validation_data=valid_data, epochs=100)

When the model is run, the error was raised

ValueError: Input 0 of layer sequential is incompatible with the layer: expected ndim=3, found ndim=0. Full shape received: []

EDIT Here is a part of csv file, world_cases -Turkey- column

0           0.0
1           0.0
2           0.0
3           0.0
4           0.0
         ...
258    291162.0
259    292878.0
260    294620.0
261    296391.0
262    298039.0

Answer 1

I reproduced your issue.

import tensorflow as tf
inputs = tf.random.normal([])
lstm = tf.keras.layers.LSTM(4)
output = lstm(inputs)
print(output.shape)

Output

ValueError: Input 0 of layer lstm_1 is incompatible with the layer: expected ndim=3, found ndim=0. Full shape received: ()

Issue is with input data shape

Since Tensorflow.Keras LSTM expects input of shape 3D. Modify your input as per this

inputs: A 3D tensor with shape [batch, timesteps, feature] 
model.add(LSTM(16,return_sequences=False)).

Working sample code

import tensorflow as tf
inputs = tf.random.normal([32, 10, 8])
lstm = tf.keras.layers.LSTM(4)
output = lstm(inputs)
print(output.shape)

Output

(32, 4)