streamlit_app.py

import streamlit as st
import tensorflow as tf
import numpy as np

# Load and prepare text
text = open("corpus/obras.txt", 'rb').read().decode(encoding='utf-8')
vocab = sorted(set(text))

char2idx = {u:i for i, u in enumerate(vocab)}
idx2char = np.array(vocab)

# Load model
model = tf.keras.models.load_model('model/modelo.h5', compile = False)

# Predictions 
def generate_text(model, start_string):
    
    num_generate = 1000

    input_eval = [char2idx[s] for s in start_string]
    input_eval = tf.expand_dims(input_eval, 0)

    text_generated = []

    temperature = 0.5

    model.reset_states()
    for i in range(num_generate):
        predictions = model(input_eval)
        
        predictions = tf.squeeze(predictions, 0)

        predictions = predictions / temperature
        
        predicted_id = tf.random.categorical(predictions, num_samples=1)[-1,0].numpy()
 
        input_eval = tf.expand_dims([predicted_id], 0)

        text_generated.append(idx2char[predicted_id])
    
    text = ''.join(text_generated)
    text = text.rsplit('.', 1)[0]

    return (start_string + text + '.')

# Streamlit

st.set_page_config(page_title= 'Gerador de Texto - Guimarães Rosa', page_icon=":eyeglasses:")

st.title('Gerador de texto de João Guimarães Rosa')
st.write('Inteligência Artficial treinada em todas as obras de João Guimarães Rosa para escrever como o autor.')

input_text = st.text_input('(Opcional) Escreva o início da frase:')

if st.button('Gerar texto'):

    if len(input_text) == 0:
        generated_text = generate_text(model, start_string = ' ')
        st.write(generated_text)
    
    else:
        generated_text = generate_text(model, start_string = input_text)
        st.write(generated_text)

with st.beta_expander("Sobre o projeto"):
    st.markdown("Todo o código do projeto está disponível no nosso repositório do [github](https://github.com/turing-usp/gerador-texto-guimaraes).")