import streamlit as st
from transformers import pipeline

# Function to get response from LLaMA 2 model
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-7b-chat-hf")
model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-2-7b-chat-hf")

def getLLamaresponse(input_text, keywords, blog_style):
    # Load the LLaMA 2 model from Hugging Face
    #model_name = model
    #llm = pipeline('text-generation', model=model_name)

    # Prompt Template
    template = """
        Generate project idea for {blog_style} by using keywords like {keywords} for the profession of {input_text}.
    """

    # Format the prompt
    prompt = template.format(blog_style=blog_style, input_text=input_text, keywords=keywords)

    # Generate the response from the LLaMA 2 model
    response = model(prompt, max_length=250, temperature=0.01)
    return response[0]['generated_text']

st.set_page_config(page_title="Generate Project Idea",
                   page_icon='🤖',
                   layout='centered',
                   initial_sidebar_state='collapsed')

st.header("Generate Project Idea 🤖")

input_text = st.text_input("Enter the Topic")

# Creating two more columns for additional fields
col1, col2 = st.columns([5, 5])

with col1:
    no_words = st.text_input('Keywords')
with col2:
    blog_style = st.selectbox('Generating project idea for',
                              ('Researchers', 'Data Scientist', 'Software Developer', 'Common People', " "), index=0)

submit = st.button("Generate")

# Final response
if submit:
    response = getLLamaresponse(input_text, no_words, blog_style)
    st.write(response)