NiyaAPI

NiyaAPI is a backend API built with Django, designed to power the Niya health platform. The API provides endpoints for medical diagnosis predictions using machine learning, as well as general-purpose AI chat and database management features. This project aims to deliver reliable, scalable, and intelligent backend services for healthcare applications.

🖥️ Frontend: Niya Frontend

---

🔷 Django Apps Overview

1. diagnosis

• Purpose: Handles all machine learning-related features, including medical diagnosis predictions based on user symptoms.
• Key Features:
  • ML model endpoints for predicting diseases
  • Utilizes pre-trained models and datasets
  • Data-driven health insights

2. core

• Purpose: Manages core backend features such as database models, AI chat, and integrations (e.g., Firebase).
• Key Features:
  • User and data management
  • AI chat endpoints (e.g., Gemini integration)
  • Admin and service utilities

---

🧭 API Routes

diagnosis App

Route	Method	Description
/diagnosis/predict/	POST	Predicts disease based on symptoms (ML model)
/diagnosis/get_prec/	GET	Get precautions for a predicted disease
/diagnosis/get_related_syms/	GET	Get related symptoms for a given symptom
/diagnosis/get_all_syms/	GET	Get all possible symptoms

core App

Route	Method	Description
/core/chat/	POST	AI chat endpoint (Gemini integration)
/core/generate_gemini_response/	POST	Generate a Gemini AI response
/core/diary_entry/	POST	Create or manage a diary entry

---

🚀 Getting Started

1. Clone the repository:

   git clone https://github.com/sabzdotpy/niyaAPI.git
   cd niyaAPI

2. Install dependencies:

   pip install -r requirements.txt

3. Run migrations:

   python manage.py migrate

4. Start the server:

   python manage.py runserver

---

⚠️ MUST DO! Environment Variables & Firebase Setup

• Environment Variables:
  1. Gemini API Key
     • Create a .env file in the project root.
     • Add Google Gemini API Key under GEMINI_API_KEY. Refer to .env.sample.
     • Example:

       GEMINI_API_KEY=your_secret_key

  2. Firebase Admin SDK
     • Go to Firebase Console > Project Settings > Service Accounts.
     • Generate a new private key and download the JSON file.
     • Add the JSON content to .env under GOOGLE_APPLICATION_CREDENTIALS_JSON.
     • Refer to .env.sample for format.

---

🗂️ Project Structure

• core/ — Core backend logic, AI chat, database models, admin, services
• diagnosis/ — ML models, prediction endpoints, health datasets
• ml/ — Machine learning scripts and saved models
• niyabackend/ — Django project settings and configuration

---

For more details, see the code and comments in each app directory.