Spaces:
Sleeping
Sleeping
A newer version of the Gradio SDK is available:
5.38.2
metadata
title: PlantIdentifier
emoji: π
colorFrom: green
colorTo: indigo
sdk: gradio
sdk_version: 5.35.0
app_file: app.py
pinned: false
πΏ Plant Identifier - Deep Learning Image Classifier
This is a deep learning-powered web app that identifies plant types from images using a Convolutional Neural Network (CNN). Built with TensorFlow and deployed using Gradio, it achieves 97% accuracy on unseen plant images.
π Live Demo on Hugging Face Spaces
π Features
- π± Identifies plant type from uploaded images
- πΈ Simple drag-and-drop image upload
- βοΈ Powered by a CNN model trained on real-world plant datasets
- π Deployed using Gradio on Hugging Face Spaces
π§ Model Overview
- Architecture: EfficientNetB0 (fine-tuned with dropout and global average pooling)
- Accuracy: ~97% on validation set
- Framework: TensorFlow / Keras
- Input Shape: 224x224 RGB images
- Dataset: Trained on the Plants Type Dataset from Kaggle
- Notebook: Full training and evaluation in:
πPlant_Image_Model_TensorFlow_Builder.ipynb
π Tech Stack
- Python
- TensorFlow / Keras
- EfficientNetB0
- Gradio (for web-based image upload UI)
- Google Colab (for model training with GPU support)
- Matplotlib / Seaborn (for evaluation visuals)
- scikit-learn (for metrics and confusion matrix)