A machine learning model that predicts prices in the Algerian market using real local market data.
Here’s a clean, solid **README.md** you can drop straight into your repo. It’s clear, recruiter-friendly, and fits a **Next.js + FastAPI** setup with `frontend/` and `backend/` folders.
---
# Algerian Market Price Prediction
A machine learning–based price prediction system built using real data from the Algerian market.
The project combines a **Next.js frontend** with a **FastAPI backend** to serve predictions through a clean web interface.
## Project Structure
```
.
├── frontend/ # Next.js application (UI)
├── backend/ # FastAPI backend (API + ML model)
└── README.md
```
## Tech Stack
### Frontend
* **Next.js**
* React
### Backend
* **FastAPI**
* Python
* Machine Learning model (for price prediction)
* Uvicorn (ASGI server)
## Features
* Predict prices based on Algerian market data
* REST API for model inference
* Modern web interface built with Next.js
* Clear separation between frontend and backend
## Getting Started
### Backend (FastAPI)
```bash
cd backend
python -m venv venv
source venv/bin/activate # Linux/macOS
# venv\Scripts\activate # Windows
pip install -r requirements.txt
uvicorn main:app --reload
```
Backend will run at:
```
127.0.0.1
```
### Frontend (Next.js)
```bash
cd frontend
npm install
npm run dev
```
Frontend will run at:
```
localhost
```
## API Documentation
Once the backend is running, you can access the interactive API docs:
```
127.0.0.1
```
## Deployment (Vercel + Render)
This project is easiest to deploy as two services:
- **Backend** (FastAPI) on **Render**
- **Frontend** (Next.js) on **Vercel**
### 1) Deploy the backend to Render
Option A (recommended): use the Blueprint in render.yaml.
1. Push this repo to GitHub.
2. In Render: **New** → **Blueprint** → select your repo.
3. Render will create a web service for the backend (`rootDir: backend`).
4. After it deploys, open the backend URL and confirm:
- `https:// .onrender.com/` returns a JSON message
- `https:/ …