📝 Handwritten Recognition API

🚀 A FastAPI-based application for handwritten recognition and image cleaning.
This repository provides a structured and scalable API setup.

---

📥 Download Model & Dataset

To run the API, download the necessary files:

1. Pretrained Model: Download Here

📂 Folder Structure

A well-organized folder structure ensures maintainability and scalability.

Directory / File	Description
app/	Main application folder
├── main.py	Initializes the FastAPI app
├── config.py	Configuration settings (environment variables)
├── routes/	API route handlers
│ ├── ocr.py	Route for image uploading
├── models/	Database models
│ ├── ocr.py	Fine-tuned model
├── utils/	Helper functions
│ ├── image.py	Image processing
tests/	Unit & integration tests
├── routes/	Tests for API endpoints
├── models/	Tests for database models
├── utils/	Tests for utility functions
.gitignore	Files to exclude from Git
requirements.txt	Dependencies for the project
README.md	Documentation

---

⚙️ Developer Workflow

Maintaining a structured development workflow ensures smooth collaboration.

🏗️ Branching Strategy

• Base Branch: dev (always pull from dev before creating a new branch)
• Branch Naming Conventions:
  • feature/<name> → For new features (e.g., feature/auth-system)
  • fix/<name> → For bug fixes (e.g., fix/user-login)
  • hotfix/<name> → For urgent fixes on production
  • refactor/<name> → For code restructuring without new features

🔄 Git Workflow

Action	Description
1️⃣ Pull Updates	git pull origin dev (Keep local dev branch updated)
2️⃣ Create Feature Branch	git checkout -b feature/<name>
3️⃣ Code & Commit	Write code, then git commit -m "Add: <feature description>"
4️⃣ Push Changes	git push origin feature/<name>
5️⃣ Open PR	Open a Pull Request (PR) to merge into dev
6️⃣ Code Review & Merge	PR is reviewed and merged into dev
7️⃣ Deploy	Once tested, merge dev into main

---

🛠️ Core Technologies

This project is built using the following stack:
(🔄 This list will be updated as more technologies are decided.)

Technology	Usage
FastAPI	Web framework for building APIs
Pydantic	Data validation and settings
Pillow (PIL)	Image handling and manipulation
Scikit-learn	Feature extraction and image processing utilities
TensorFlow / PyTorch	Deep learning models for handwritten text recognition

---

🔖 Best Practices 🚀

✅ Always create feature branches from `dev`

✅ Use clear commit messages

✅ Add: User authentication system

✅ Pull the latest changes before pushing

✅ git pull origin dev

✅ Write unit tests for all new features

✅ Code reviews before merging PRs

✅ git pull origin dev