File size: 826 Bytes
9fe323f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 |
import streamlit as st
from transformers import T5ForConditionalGeneration, T5Tokenizer
model = T5ForConditionalGeneration.from_pretrained("t5-small")
tokenizer = T5Tokenizer.from_pretrained("t5-small")
def generate_response(input_text):
input_ids = tokenizer.encode("chatbot: " + input_text, return_tensors="pt", max_length=512)
output_ids = model.generate(input_ids, max_length=100, num_beams=1, early_stopping=True)
response = tokenizer.decode(output_ids[0], skip_special_tokens=True)
return response
# Set up Streamlit app
st.title("Simple Chatbot with T5")
user_input = st.text_input("You:", "")
if st.button("Send"):
if user_input.strip() != "":
response = generate_response(user_input)
st.text_area("Bot:", response)
else:
st.warning("Please enter a valid input.")
|