🌤️ Weather GPT Assistant

A smart weather assistant powered by GPT Function Calling and OpenWeather API.
It suggests what to wear, do, and prepare for — all based on real-time and forecasted weather data.

🚀 Features

• ✨ GPT with Function Calling
• 🌍 Real-time & 7-day weather forecasts
• 👕 Outfit suggestions with emojis
• 🏞️ Activity planning tips
• ⚠️ Weather warnings and preparation advice
• 🖥️ Streamlit-based UI with selectable assistant variants (default, emoji, detailed)

🎥 Demo

🧠 Tech Stack

• LLM: OpenAI GPT (gpt-4 or gpt-4o)
• Weather API: Weather API (current & forecast)
• Frontend: Streamlit
• Backend: Python, OpenAI Function Calling
• Logging: Python logging module
• Structure:

  .
  ├── assets/                # Images, icons or static files
  ├── logs/                  # Application logs
  ├── scripts/               # CLI scripts for quality checks or launching
  ├── src/
  │   ├── services/          # Service-level logic (TBD)
  │   ├── tools/             # Tool calling: weather fetch, GPT call
  │   └── gpt_call.py        # GPT call
  ├── utils/                 # Constants, formatter, logging, prompt initializer
  ├── .env.example           # Sample environment config
  ├── app.py                 # Streamlit entrypoint
  └── README.md
  

⚙️ Installation

# 1) Virtual environment
python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate

# 2) Dependencies
pip install -r requirements.txt

# 3) Environment variables (optional but recommended)
cp .env.example .env
# Add your OPENAI_API_KEY and WEATHER_API_KEY to .env

# 4) Run the app
streamlit run app.py
# or
bash scripts/run.sh

🔑 Environment Variables

From .env.example:

OPENAI_API_KEY=your_api_key_here
DEFAULT_MODEL=gpt-4o-mini

# Logging (optional)
LOG_LEVEL=INFO
LOG_TO_CONSOLE=1
LOG_FILE=app.log
LOG_MAX_BYTES=1000000
LOG_BACKUPS=3
LOG_FORMAT=%(asctime)s | %(levelname)s | %(name)s | %(message)s
LOG_DATEFMT=%Y-%m-%d %H:%M:%S
WEATHER_API_KEY=your_weather_api_key

• Logging configuration is fully adjustable via environment variables.

🧪 Quality & Tests

# One command for all checks:
bash scripts/quality.sh

# Or manually:
ruff check . --fix
black . --line-length 100
pylint $(git ls-files '*.py' | tr '\n' ' ')
pytest -q

ruff → fast linter, black → formatting, pylint → deep analysis, pytest → tests.

📸 UI

Variant	Example
Emoji 😎	Suggests outfit + activity + warning with icons
Default 💬	Text-based assistant style
Detailed 📚	Full explanation and safety tips
Short 🧾	Gives concise outfit and activity suggestions

🧩 You can choose both the OpenAI model and the assistant variant (e.g., Emoji, Detailed, Short) from the sidebar.

The assistant will respond accordingly based on your selections.

🧹 Clear Conversation to reset the assistant’s memory and start fresh

📄 License

This project is licensed under the MIT License.