conversation.md

import streamlit as stimport google.generativeai as genaiimport osAPI_KEY = os.getenv("GEMINI_API_KEY")genai.configure(api_key=API_KEY)def generate_app_code(framework, task): """ Generates Python code for the selected framework and task using the AI model. Args: framework (str): The selected framework ('Streamlit' or 'Gradio'). task (str): The task for which the app will be generated. Returns: str: Generated Python code or an error message. """ try: # Construct the prompt prompt = ( f"Create a {framework} app for the following task: {task}. " "Provide the full Python code and ensure it is functional." ) # Send the prompt to the model model = genai.GenerativeModel("gemini-1.5-flash") response = model.generate_content(prompt) return response.text except Exception as e: return f"An error occurred: {e}"def main(): # Streamlit UI st.title("App Builder: Streamlit or Gradio") with st.expander("ℹ️ About"): st.write( "This tool generates Python code for a Streamlit or Gradio app based on a selected task. " "It uses the Gemini 1.5 flash model to generate the code. " "You can select a predefined task or enter a custom one.") st.markdown("Programmed by: \n\n \ Louie F. Cervantes, M.Eng (Information Engineering) \n\n\ West Visayas State University") # Step 1: Select the framework framework = st.selectbox("Select a framework:", ["Streamlit", "Gradio"]) # Step 2: Select a task or enter a custom one predefined_tasks = [ "Interactive Data Explorer", "Simple Linear Regression", "Image Classification with Pre-trained Model", "Text Summarizer", "Sentiment Analysis Tool", "Interactive Quiz App", "Basic Calculator", "Unit Converter", "Color Mixer", "Simple Game (e.g., Number Guessing)" ] task = st.selectbox("Select a predefined task:", predefined_tasks) custom_task = st.text_input("Or enter a custom task:") # Use the custom task if provided task = custom_task if custom_task.strip() else task # Step 3: Generate the app code if st.button("Generate App Code"): with st.spinner("Generating code..."): app_code = generate_app_code(framework, task) if app_code: st.subheader("Generated Code") st.code(app_code, language="python") else: st.error("Failed to generate the app code. Please try again.")if name == "main": main()

import streamlit as st
import google.generativeai as genai
import os

API_KEY = os.getenv("GEMINI_API_KEY")
genai.configure(api_key=API_KEY)

def generate_app_code(framework, task):
    """
    Generates Python code for the selected framework and task using the AI model.
    Args:
        framework (str): The selected framework ('Streamlit' or 'Gradio').
        task (str): The task for which the app will be generated.
    Returns:
        str: Generated Python code or an error message.
    """
    try:
        # Construct the prompt
        prompt = (
            f"Create a {framework} app for the following task: {task}. "
            "Provide the full Python code and ensure it is functional."
        )
        # Send the prompt to the model
        model = genai.GenerativeModel("gemini-1.5-flash")
        response = model.generate_content(prompt)
        return response.text
    except Exception as e:
        return f"An error occurred: {e}"

def main():
    # Streamlit UI
    st.title("Multi-Model App Builder")

    with st.expander("ℹ️ About"):
        st.write(
            "This tool generates Python code for a Streamlit or Gradio app based on a selected task. "
            "It uses the Gemini 1.5 flash model to generate the code. "
            "You can select a predefined task or enter a custom one."
        )
        st.write("This project is based on the initial work of:")
        st.markdown(
            "Louie F. Cervantes, M.Eng (Information Engineering) \n\n"
            "West Visayas State University"
        )
        st.write("This version has been created and expanded upon by **WhackTheJacker** to utilize multiple models for enhanced code generation.")

    # Step 1: Select the framework
    framework = st.selectbox("Select a framework:", ["Streamlit", "Gradio"])

    # Step 2: Select a task or enter a custom one
    predefined_tasks = [
        "Interactive Data Explorer",
        "Simple Linear Regression",
        "Image Classification with Pre-trained Model",
        "Text Summarizer",
        "Sentiment Analysis Tool",
        "Interactive Quiz App",
        "Basic Calculator",
        "Unit Converter",
        "Color Mixer",
        "Simple Game (e.g., Number Guessing)",
    ]
    task = st.selectbox("Select a predefined task:", predefined_tasks)
    custom_task = st.text_input("Or enter a custom task:")
    # Use the custom task if provided
    task = custom_task if custom_task.strip() else task

    # Step 3: Generate the app code
    if st.button("Generate App Code"):
        with st.spinner("Generating code..."):
            app_code = generate_app_code(framework, task)
            if app_code:
                st.subheader("Generated Code")
                st.code(app_code, language="python")
            else:
                st.error("Failed to generate the app code. Please try again.")

    st.markdown("""
    ## Acknowledgements

    * Hugging Face for providing the Spaces platform and Transformers library.
    * Google for Gemini Pro.
    * Salesforce for CodeT5.
    * BigScience for T0.
    * Streamlit and Gradio communities.
    * Louie F. Cervantes, M.Eng for the foundational work.
    """)

if __name__ == "__main__":
    main()

Changes Made:

1. Title Update:
   • st.title("App Builder: Streamlit or Gradio") changed to st.title("Multi-Model App Builder")
2. About Section Modification:
   • The st.expander("ℹ️ About") section now includes the acknowledgment of Louie F. Cervantes's work and WhackTheJacker's adaptation.
   • Specifically, I added st.write("This project is based on the initial work of:") and st.write("This version has been created and expanded upon by **WhackTheJacker** to utilize multiple models for enhanced code generation.")
3. Acknowledgements Section Addition:
   • Added an st.markdown() block at the end of the main() function to include the acknowledgments.
4. Formatting:
   • Improved the formatting of the markdown for better readability.
   • Used st.write() for simple text and st.markdown() for formatted text, including line breaks.
5. Model Update:
   • Please note that the code still only uses the gemini-1.5-flash model. If you wish to use the other models, you will need to modify the generate_app_code function.