如何为 LSTM keras 重塑 X_train 和 y_train

Question

I have the following:

X_train.shape
(2730, 10)

y_train.shape
(2730)

I want to train LSTM model with keras, but I'm not sure how to reshape the input.

I have added this LSTM layer

time_steps = 30
input_dim = 10 # number of features
...
self.model.add(LSTM(self.hidden_dim, input_shape=(time_steps, self.input_dim), return_sequences=True))
...

The input_shape doesn't match my input. How should I reshape my X_train? Do I also have to reshape the y_train?

Answer 1

How should I reshape my X_train?

The simplest option would be to add a timesteps dimension to your data to make it compatible with an LSTM :

import tensorflow as tf

samples = 5
features = 10
data = tf.random.normal((samples, features))
time_series_data = tf.expand_dims(data, axis=1) # add timesteps dimension
tf.print('Data -->', tf.shape(data), 'Time series data', tf.shape(time_series_data))
# Data --> [5 10] Time series data [5 1 10]

However, you want to use 30 timesteps for each feature leading to the shape (samples, 30, 10) . So, what you can use is the RepeatVector layer as part of your model or tf.repeat . Here is an example with the RepeatVector layer:

model = tf.keras.Sequential()
model.add(tf.keras.layers.Dense(10, input_shape=(features,)))
model.add(tf.keras.layers.RepeatVector(30))
model.add(tf.keras.layers.LSTM(32))
model.add(tf.keras.layers.Dense(1, activation='sigmoid'))
model.build((1, 10))
tf.print(model.summary())

Model: "sequential_01"
_________________________________________________________________
 Layer (type)                Output Shape              Param #   
=================================================================
 dense_24 (Dense)            (None, 10)                110       
                                                                 
 repeat_vector_1 (RepeatVect  (None, 30, 10)           0         
 or)                                                             
                                                                 
 lstm_3 (LSTM)               (None, 32)                5504      
                                                                 
 dense_25 (Dense)            (None, 1)                 33        
                                                                 
=================================================================
Total params: 5,647
Trainable params: 5,647
Non-trainable params: 0
_________________________________________________________________
None

You could also first map the 10 features to a 300-dimensional output and then reshape the output to fit into an LSTM :

model = tf.keras.Sequential()
model.add(tf.keras.layers.Dense(300, input_shape=(features,)))
model.add(tf.keras.layers.Reshape((30, 10)))
model.add(tf.keras.layers.LSTM(32))
model.add(tf.keras.layers.Dense(1, activation='sigmoid'))

Model: "sequential_02"
_________________________________________________________________
 Layer (type)                Output Shape              Param #   
=================================================================
 dense_26 (Dense)            (None, 300)               3300      
                                                                 
 reshape (Reshape)           (None, 30, 10)            0         
                                                                 
 lstm_4 (LSTM)               (None, 32)                5504      
                                                                 
 dense_27 (Dense)            (None, 1)                 33        
                                                                 
=================================================================
Total params: 8,837
Trainable params: 8,837
Non-trainable params: 0
_________________________________________________________________
None

To the question:

Do I also have to reshape the y_train?

It depends on what you want. If you only have a simple classification task, as I have assumed in the examples, then you do not need to change y_train.