import streamlit as st
from langchain.prompts import PromptTemplate
from langchain.llms import CTransformers

def getLLamaresponse(input_text, no_words, blog_style):
    # Initialize the language model with the specified configuration
    # Use a pipeline as a high-level helper
    from transformers import pipeline
    llm = pipeline("text-generation", model="TheBloke/Llama-2-7B-Chat-GGML")
    # Define the template for the prompt
    template = """
    Write a blog in {blog_style} job profile for the topic "{input_text}" within {no_words} words.
    """
    
    # Create the prompt using the template
    prompt = PromptTemplate(input_variables=['blog_style', 'input_text', 'no_words'], template=template)
    
    # Format the prompt with the provided parameters
    formatted_prompt = prompt.format(blog_style=blog_style, input_text=input_text, no_words=no_words)
    
    # Generate the response using the language model
    response = llm(formatted_prompt)
    
    return response

# Configure the Streamlit app
st.set_page_config(page_title="Generate Blog",
                   layout='centered',
                   initial_sidebar_state='collapsed')

# App header
st.header("Generate Blog")

# Input for the blog topic
input_text = st.text_input("Enter the blog topic")

# Columns for additional inputs
col1, col2 = st.columns([5, 5])
with col1:
    no_words = st.text_input('Number of words')

with col2:
    blog_style = st.selectbox('Writing the blog for', ('researchers', 'data scientists', 'common people'), index=0)

# Generate button
submit = st.button("Generate")

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