app.py

import streamlit as st
import numpy as np
import pandas as pd
import os
import torch
import torch.nn as nn
from transformers import AutoTokenizer, AutoModelWithLMHead, AutoModelForCausalLM
from transformers.activations import get_activation


st.title('Informal to Formal:')
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

st.text('''Check out this other space: https://huggingface.co/spaces/BigSalmon/GPT2Space''')

st.text('''How To Make Prompt: https://huggingface.co/BigSalmon/DefinitionsSynonyms3

part of speech- verb
definition: grow less in intensity or degree
ex. rather than leave immediately and be drenched, they waited for the storm to ________
synonyms: subside; moderate; decrease
antonyms: increase
word: abate''')

@st.cache(allow_output_mutation=True)
def get_model():
    #tokenizer = AutoTokenizer.from_pretrained("gpt2")
    #model = AutoModelWithLMHead.from_pretrained("BigSalmon/MrLincoln12")
    #model = AutoModelWithLMHead.from_pretrained("BigSalmon/Points")
    #tokenizer = AutoTokenizer.from_pretrained("BigSalmon/InformalToFormalLincoln91Paraphrase")
    #model = AutoModelForCausalLM.from_pretrained("BigSalmon/InformalToFormalLincoln91Paraphrase")
    #tokenizer = AutoTokenizer.from_pretrained("BigSalmon/InformalToFormalLincoln95Paraphrase")
    #model = AutoModelForCausalLM.from_pretrained("BigSalmon/InformalToFormalLincoln95Paraphrase")
    tokenizer = AutoTokenizer.from_pretrained("BigSalmon/DefinitionsSynonyms3")
    model = AutoModelForCausalLM.from_pretrained("BigSalmon/DefinitionsSynonyms3")
    return model, tokenizer
    
model, tokenizer = get_model()

with st.form(key='my_form'):
    prompt = st.text_area(label='Enter sentence')
    submit_button = st.form_submit_button(label='Submit')

    if submit_button:
      with torch.no_grad():
        text = tokenizer.encode(prompt)
        myinput, past_key_values = torch.tensor([text]), None
        myinput = myinput
        myinput= myinput.to(device)
        logits, past_key_values = model(myinput, past_key_values = past_key_values, return_dict=False)
        logits = logits[0,-1]
        probabilities = torch.nn.functional.softmax(logits)
        best_logits, best_indices = logits.topk(100)
        best_words = [tokenizer.decode([idx.item()]) for idx in best_indices]
        text.append(best_indices[0].item())
        best_probabilities = probabilities[best_indices].tolist()
        words = []              
        st.write(best_words)