Apppublic

ehsanulhaque92/house-plant-api

sourceHugging Faceupdated 1y agoView on Hugging Face

0likes

App README

🌿 House Plant Species Identifier API

This repository contains the containerized Flask backend for the House Plant Species Identifier, a full-stack, decoupled machine learning application.

Project Overview

This API serves a fine-tuned PyTorch (EfficientNet-B0) model capable of identifying 47 different species of house plants. It receives an image file via a POST request to the /predict endpoint and returns the predicted species and a confidence score in JSON format.

This backend is designed to be a standalone microservice, containerized with Docker, and is currently deployed on Hugging Face Spaces.

🔗 Project Links

Link	URL
🚀 Live Demo	[house-plant-frontend.vercel.app](https://house-plant-frontend-3vsr32tzq-md-ehsanul-haque-kanans-projects.vercel.app/)
🎨 Frontend Repository (React)	github.com/MdEhsanulHaqueKanan/house-plant-frontend
⚙️ Backend API Repository (This Repo)	github.com/MdEhsanulHaqueKanan/house-plant-api
📦 Original Monolithic Project	github.com/MdEhsanulHaqueKanan/house-plant-species-identifier-machine-learning-flask-app

Technology Stack

—Backend: Python, Flask
—Machine Learning: PyTorch, Torchvision, Pillow
—Containerization: Docker
—Deployment: Hugging Face Spaces