Home
AI/ML Projects
Machine Learning Project

Succulent Classifier

A lightweight computer vision model that identifies succulents with 97% accuracy.
Share this post

Overview

I’ve always liked plants, especially succulents, but I can never seem to remember their names. So I built a machine learning project that could classify them and help me identify what I’m looking at.

I created a lightweight computer vision model using Fastai to recognize three common types: Echeveria, Agave, and Cactus. I then deployed the model with Gradio and hosted it on Hugging Face Spaces. This project gave me hands-on experience with the full ML deployment process, from training to building a user-friendly interface.

Hugging Face Interface for Plant type model

Key Highlights

  • Trained an image classifier using transfer learning with ResNet18
  • Reached 97% accuracy in identifying Echeveria correctly
  • Designed a clean, user-friendly interface using Gradio
  • Made the demo accessible to anyone through Hugging Face
Share this post
Design
Product
Research