LSTM model building

LSTM-buliding.py

### Create the Stacked LSTM model
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense,Dropout
from tensorflow.keras.layers import LSTM
from tensorflow.keras.initializers import RandomNormal, Constant
import tensorflow as tf
# Set random seed for reproducibility: get the same result after each time running the model
tf.random.set_seed(1234)


# # Build the LSTM Stack model
model=Sequential()
# Adding first LSTM layer
model.add(LSTM(150,return_sequences=True,input_shape=(100,1)))
model.add(Dropout(0.2)) # Dropout regularisation
# second LSTM layer 
model.add(LSTM(150,return_sequences=True))
# Adding third LSTM layer 
model.add(LSTM(150, return_sequences=True))
model.add(Dropout(0.2))
# Adding fourth LSTM layer
model.add(LSTM(150)) #  you can try 100 as well
model.add(Dropout(0.2))
# Adding the Output Layer
model.add(Dense(1))
model.compile(loss='mean_squared_error',optimizer='adam')




# summary of the model  
model.summary()

## Output:
# Model: "sequential_8"
# _________________________________________________________________
#  Layer (type)                Output Shape              Param #   
# =================================================================
#  lstm_32 (LSTM)              (None, 100, 150)          91200     
                                                                 
#  dropout_26 (Dropout)        (None, 100, 150)          0         
                                                                 
#  lstm_33 (LSTM)              (None, 100, 150)          180600    
                                                                 
#  lstm_34 (LSTM)              (None, 100, 150)          180600    
                                                                 
#  dropout_27 (Dropout)        (None, 100, 150)          0         
                                                                 
#  lstm_35 (LSTM)              (None, 150)               180600    
                                                                 
#  dropout_28 (Dropout)        (None, 150)               0         
                                                                 
#  dense_8 (Dense)             (None, 1)                 151       
                                                                 
# =================================================================
# Total params: 633,151
# Trainable params: 633,151
# Non-trainable params: 0
# _________________________________________________________________

# use early stop to avoid over-fiting 
from tensorflow.keras.callbacks import EarlyStopping

monitor = EarlyStopping(monitor='val_loss', min_delta=1e-3, patience=30, 
        verbose=1, mode='auto', restore_best_weights=True)

history=model.fit(X_train,y_train,validation_data=(X_test,ytest),
        callbacks=[monitor],verbose=1,epochs=1000)


#history= model.fit(X_train,y_train,validation_data=(X_test,ytest),epochs=54,batch_size=64,verbose=1)