|
|
--- |
|
|
title: CIFAR-10 Image Classifier |
|
|
emoji: 🖼️ |
|
|
colorFrom: blue |
|
|
colorTo: green |
|
|
sdk: streamlit |
|
|
sdk_version: 1.52.0 |
|
|
app_file: app.py |
|
|
pinned: false |
|
|
--- |
|
|
|
|
|
# 🚀 CIFAR-10 Image Classifier |
|
|
|
|
|
This is a **Streamlit** app running a **TensorFlow/Keras CNN** trained on the CIFAR-10 dataset. |
|
|
Upload an image, and the model will predict one of 10 classes: |
|
|
|
|
|
**airplane, automobile, bird, cat, deer, dog, frog, horse, ship, truck** |
|
|
|
|
|
--- |
|
|
|
|
|
## 🧩 Features |
|
|
- Upload images (PNG/JPG) |
|
|
- Auto-resize and preprocess |
|
|
- Predict top-3 classes with confidence |
|
|
- Visualize all class probabilities in a bar chart |
|
|
- Works directly on Hugging Face Spaces |
|
|
|
|
|
--- |
|
|
|
|
|
## 📁 Files |
|
|
- `app.py` — The Streamlit app |
|
|
- `models/cifar10_cnn.keras` — Trained model |
|
|
- `models/labels_map.json` — Class labels |
|
|
- `requirements.txt` — Dependencies |
|
|
|
|
|
--- |
|
|
|
|
|
## ⚡ How to run |
|
|
This app is fully deployed on **Hugging Face Spaces**. Just push your repo with this structure, and it will build automatically. |
