README.md

metadata

title: CIFAR-100 ResNet18 Demo
emoji: 🖼️
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 4.44.0
app_file: app.py
pinned: false
license: mit
models:
  - ivantv/cifar100-resnet18

CIFAR-100 Image Classification Demo

This is a demo Space for the ResNet-18 model trained on CIFAR-100 dataset.

Model: ivantv/cifar100-resnet18

Features

• 🎯 100 Categories - Classify images into animals, vehicles, objects, plants, and more
• 🚀 75.84% Accuracy - High-performance ResNet-18 model
• 📸 Easy to Use - Just upload an image and get instant predictions
• 🔝 Top-5 Predictions - See the most likely classes with confidence scores

Model Details

• Architecture: ResNet-18
• Parameters: 11,220,132
• Training: 50 epochs on CIFAR-100
• Test Accuracy: 75.84%

Usage

1. Upload an image using the interface
2. View the top-5 predictions with confidence scores
3. Try different types of images!

Categories

The model can recognize 100 different categories including:

Animals: bear, tiger, elephant, dolphin, fox, lion, wolf, butterfly, etc.
Vehicles: bicycle, bus, train, motorcycle, pickup_truck, etc.
Objects: chair, table, lamp, clock, keyboard, telephone, etc.
Plants: maple_tree, oak_tree, palm_tree, rose, tulip, orchid, etc.
Structures: house, castle, bridge, skyscraper, etc.

Files

This Space uses:

• app.py - Gradio interface
• requirements.txt - Dependencies
• Model downloaded from: ivantv/cifar100-resnet18

License

MIT License