LSTM_model.py

# LSTM_model.py
import numpy as np
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Embedding, LSTM, Dense, Dropout
from data_preprocessing import preprocess_data, split_data
import joblib  # To save the tokenizer and label encoder

# Define the LSTM model
def build_lstm_model(vocab_size, embedding_dim=64, max_len=10, lstm_units=128, dropout_rate=0.2, output_units=6):
    model = Sequential()
    model.add(Embedding(input_dim=vocab_size, output_dim=embedding_dim, input_length=max_len))
    model.add(LSTM(units=lstm_units, return_sequences=False))
    model.add(Dropout(dropout_rate))
    model.add(Dense(units=output_units, activation='softmax'))
    
    model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
    
    return model

# Main function to execute the training process
def main():
    # Path to your data file
    data_path = r"E:\transactify\transactify\transactify\transactify\transactify\data_set\transaction_data.csv"

    # Preprocess the data
    sequences, labels, tokenizer, label_encoder = preprocess_data(data_path)

    # Check if preprocessing succeeded
    if sequences is not None:
        print("Data preprocessing successful!")
        
        # Split the data into training and testing sets
        X_train, X_test, y_train, y_test = split_data(sequences, labels)
        print(f"Training data shape: {X_train.shape}, Training labels shape: {y_train.shape}")
        print(f"Testing data shape: {X_test.shape}, Testing labels shape: {y_test.shape}")

        # Build the LSTM model
        vocab_size = tokenizer.num_words + 1  # +1 for padding token
        model = build_lstm_model(vocab_size, max_len=10, output_units=len(label_encoder.classes_))

        # Train the model
        model.fit(X_train, y_train, epochs=50, batch_size=8, validation_data=(X_test, y_test))

        # Evaluate the model
        loss, accuracy = model.evaluate(X_test, y_test)
        print(f"Test Loss: {loss:.4f}, Test Accuracy: {accuracy:.4f}")

        # Save the model
        model.save('transactify.h5')
        print("Model saved as 'transactify.h5'")

        # Save the tokenizer and label encoder
        joblib.dump(tokenizer, 'tokenizer.joblib')
        joblib.dump(label_encoder, 'label_encoder.joblib')
        print("Tokenizer and LabelEncoder saved as 'tokenizer.joblib' and 'label_encoder.joblib'")

    else:
        print("Data preprocessing failed.")

# Execute the main function
if __name__ == "__main__":
    main()