from pathlib import Path
from sklearn.model_selection import train_test_split
import torch
from torch.utils.data import Dataset
from transformers import DistilBertTokenizerFast, DistilBertForSequenceClassification
from transformers import Trainer, TrainingArguments

import streamlit as st
from streamlit_chat import message
import requests

 model_one = "distilbert-base-uncased-finetuned-sst-2-english"
 model_two = "Newtral/xlm-r-finetuned-toxic-political-tweets-es"
 
def toxicRating(text, model):
    model = AutoModelForSequenceClassification.from_pretrained(model)
    tokenizer = AutoTokenizer.from_pretrained(model)
    
    classifier = pipeline("sentiment-analysis", model=model, tokenizer=tokenizer)
    results = classifier(text)
    return results
    
def main():
    st.title("TOXIC TWEETS, \n TOXIC OR NOT?")
    prompt = st.header("Select Model")
    selection = st.radio("Models",('Model 1', 'Model 2'))
                         
    
    input = st.text_area("Enter Tweet: ")
        if input:
            if selection == 'Model 1':
                rating = rate_ModelOne(input, model_one)
                st.write(f"Label: {rating[1]} \n Score  : {rating[3]}")
            elif selection == 'Model 2':
                rating = rate_ModelTwo(input, model_two)
                rating = rate_ModelOne(input, model_one)
                st.write(f"Label: {rating[1]} \n Score  : {rating[3]}")
            else:
                st.warning("Enter Tweet")

if __name__ == "__main__":
    main();