import streamlit as st
import pandas as pd
import nltk
import pickle
from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize
import tensorflow as tf
from tensorflow.keras.preprocessing.text import Tokenizer
from tensorflow.keras.preprocessing.sequence import pad_sequences
from tensorflow.keras.models import load_model

nltk.download('punkt')
nltk.download('stopwords')

stop_words = set(stopwords.words('english'))

vocab_size = 50000
embedding_dim = 100
max_length = 50

model_100epoch = load_model("test.h5")
tokenizer_filename = "tokenizertest.pkl"
with open(tokenizer_filename, 'rb') as tokenizer_file:
    loaded_tokenizer = pickle.load(tokenizer_file)

def preprocess_text(text):
    if type(text) == str:
        tokens = word_tokenize(text.lower())
        filtered_tokens = [word for word in tokens if word.isalnum() and word not in stop_words]
        return ' '.join(filtered_tokens)
    else:
        return str(text)


def predict_sentiment(tweet,model):
    preprocessed_tweet = preprocess_text(tweet)
    tweet_sequence = loaded_tokenizer.texts_to_sequences([preprocessed_tweet])
    padded_sequence = pad_sequences(tweet_sequence, maxlen=max_length, padding='post', truncating='post')
    sentiment_prediction = model.predict(padded_sequence)[0][0]
    return sentiment_prediction

def result(prediction):
    if(prediction>0.7445957064628601):
        return "positive"
    elif(prediction>-0.4008907079696655):
        return "neutral"
    else:
        return "negative"


tweet = st.text_input('Enter a tweet related to the political situation in India.')
sentiment_prediction = predict_sentiment(tweet,model_100epoch)
st.write("Sentiment prediction in relation to Mr. Narendra Modi  :", sentiment_prediction, " => ",result(sentiment_prediction)," sentiment.")