pages/1_Simple_Chat_UI.py

import streamlit as st
from transformers import AutoTokenizer, AutoModelForCausalLM, AutoModelForSeq2SeqLM

def load_model_tokenizer(model_name, hf_api_key):
    if model_name == "LLaMa-2B":
        model_name="llmware/bling-sheared-llama-2.7b-0.1"
        model = AutoModelForCausalLM.from_pretrained(model_name, token=hf_api_key)
        tokenizer = AutoTokenizer.from_pretrained(model_name, tokenizer=hf_api_key)
    elif model_name == "Red-Pajamas-3b":
        model_name = "llmware/bling-red-pajamas-3b-0.1"
        model = AutoModelForCausalLM.from_pretrained(model_name)
        tokenizer = AutoTokenizer.from_pretrained(model_name)
    return (model,tokenizer)

def generate_response(prompt_input, model, tokenizer):
    inputs = tokenizer.encode_plus(prompt_input, return_tensors="pt")
    # Generate the response from the model with additional parameters
    outputs = model.generate(**inputs, max_length=max_length, do_sample=True ,temperature=temperature)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True).strip()
    return response

st.set_page_config(page_title="Learn Geoscience")

with st.sidebar:
    st.title('Learn Geoscience Chat')
    if 'hf_key' in st.secrets:
        st.success('Huggingface API key provided', icon='✅')
        hf_api_key = st.secrets['hf_key']
    else:
        hf_api_key = st.text_input('Enter Huggingface API Key:', type='password')
        if not hf_api_key:
            st.warning('Please enter Huggingface API key!', icon='⚠️')
        else:
            st.success('Proceed to entering your prompt message!', icon='👉')
    max_length = st.slider("Max Length", 10, 100, 50)
    temperature = st.slider("Temperature", 0.0, 1.0, 0.7)

if "messages" not in st.session_state.keys():
    st.session_state.messages = [{"role": "assistant", "content": "How may I help you?"}]

model_name = st.radio("Select model to chat", options=["LLaMa-2B", "Red-Pajamas-3b"], horizontal=True, key='model_selection')
model, tokenizer = load_model_tokenizer(model_name, hf_api_key)

for message in st.session_state.messages:
    with st.chat_message(message["role"]):
        st.write(message["content"])

if prompt := st.chat_input(disabled = not hf_api_key):
    st.session_state.messages.append({"role": "user", "content": prompt})
    with st.chat_message("user"):
        st.write(prompt)

if st.session_state.messages[-1]["role"] != "assistant":
    with st.chat_message("assistant"):
        with st.spinner("Thinking..."):
            response = generate_response(prompt, model, tokenizer)
            st.write(response)
    message = {"role": "assistant", "content": response}
    st.session_state.messages.append(message)