main.py

import streamlit as st
import numpy as np
from tensorflow.keras.models import load_model
from tensorflow.keras.preprocessing.sequence import pad_sequences
import pickle
model = load_model("next_word_lstm.h5")

with open("tokenizer.pickle","rb") as file:
    tokenizer = pickle.load(file)
 
 
def predict_next_word(model,tokenizer,text,max_sequence_len):
    token_list = tokenizer.texts_to_sequences([text])[0]
    if len(token_list) >= max_sequence_len:
        token_list = token_list[-(max_sequence_len-1):]
    token_list = pad_sequences([token_list],maxlen = max_sequence_len-1,padding="pre")
    predicted = model.predict(token_list, verbose = 0) 
    predict_next_index = np.argmax(predicted,axis=1)
    for word,index in tokenizer.word_index.items():
        if index == predict_next_index:
            return word    
    return None      


st.title("Next word Prediction With LSTM RNN and early stopping")
input_text = st.text_input("Enter a sequence of words, eg : Their was a boy who")
if st.button("Predict Next Word"):
     max_sequence_len = model.input_shape[1] + 1
     next_word = predict_next_word(model,tokenizer,input_text,max_sequence_len)
     st.write(f"Next Word : {next_word}")