Task Planner

About

I built this project to fulfill the requirements for the COMP6065001 - LEC course at Binus University.

Task Planner is a scheduling tool that uses a mix of large language models (Google's Gemini Flash 2.0 Exp) and a Random Forest machine learning model. Instead of just blocking out time on a calendar, it tries to allocate your tasks based on complexity, available hours, and energy levels.

How it works

The core scheduler sits in src/schedule_ai.py. It takes your high-level goals and time constraints, runs them through the ML model to guess the complexity, and then passes everything to Gemini to generate a readable, prioritized schedule.

We use Gradio for the frontend, which gives you a simple web interface to input your constraints and see the output immediately.

Getting started

You need Python 3.9 or newer.

1. Clone the repository:

git clone https://github.com/ghtmarco/Task-Planner.git
cd Task-Planner

2. Install the requirements:

pip install -r requirements.txt

3. Set up your environment variables. You will need a Google Gemini API key. Copy the example file and add your key:

cp .env.example .env

4. Run the app:

python app.py

The Gradio interface will start locally. Open the provided localhost link in your browser to start generating schedules.

Using it in code

If you prefer to bypass the UI and use the scheduler directly in your own scripts:

from src.schedule_ai import SimpleScheduler

planner = SimpleScheduler()

schedule = planner.generate_schedule(
    duration="weekly",
    goals="Learn AI fundamentals and practice coding",
    available_hours=4,
    considerations="Morning study sessions preferred"
)

print(schedule)

Repository structure

• app.py: The Gradio web interface.
• src/schedule_ai.py: Core scheduling logic and API calls.
• Models/: Contains the trained Random Forest model (random_forest_model.pkl) and scaler (scaler.pkl).
• docs/: Contributing guidelines and environment variable documentation.

License

This project is licensed under the MIT License. See LICENSE for details.