Octopi ML API

📌 Project Description

Octopi ML API is a machine learning-powered backend designed to predict software development effort. The model estimates the effort required (in person-hours) based on various project metrics such as Adjusted Function Points (AFP), transaction counts, and code churn.

This AI model was trained on a comprehensive China dataset containing over 500 rows of software projects, ensuring robust and data-driven predictions.

The API exposes endpoints to interact with the trained model and retrieve feature information.

🔗 Frontend Repository: Octopi Frontend

🛠️ Technologies Used

This project leverages the following technologies and libraries:

Core Frameworks

• Flask: A lightweight WSGI web application framework.
• Flask-RESTx: An extension for Flask for fast building of REST APIs with Swagger documentation.

Machine Learning & Data Processing

• Scikit-learn: The machine learning library used for training the prediction model.
• Pandas: For data manipulation and analysis.
• NumPy: For numerical computing.
• SciPy: For scientific computing and technical computing.
• Joblib: For efficient saving and loading of the trained model.

Deployment & Tools

• Railway: Cloud platform used for deployment.
• Gunicorn: WSGI HTTP Server for UNIX.

🚀 API Endpoints

The API provides the following endpoints. You can also explore the interactive Swagger documentation at /docs when running the app.

Method	Endpoint	Description
POST	/predict/	Predicts effort based on project metrics.
GET	/predict/health	Checks the API health status.
GET	/predict/features	Returns a list of required input features.

Example Prediction Request

{
  "AFP": 1587,
  "Input": 774,
  "Output": 260,
  "Enquiry": 340,
  "File": 128,
  "Interface": 0,
  "Duration": 4,
  "Resource": 4,
  "Added": 1502,
  "Changed": 0,
  "Deleted": 0,
  "PDR_AFP": 4.7
}

📦 Installation

1. Clone the repository:

   git clone https://github.com/seif-a096/Octopi_model_api.git
   cd Octopi_model_api

2. Install dependencies:

   pip install -r requirements.txt

3. Run the application:

   python app.py

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.