mishl/roxy-classifier

Roxy Migurdia Character Classifier

A YOLO-based image classifier trained to recognize Roxy Migurdia from the anime/manga series "Mushoku Tensei: Jobless Reincarnation".

Model Details

• Task: Image Classification
• Input Size: 224x224 pixels
• Classes: Roxy Migurdia detection
• Framework: Ultralytics

Training Configuration

• Epochs: 100
• Batch Size: 64
• Optimizer: Auto (SGD with momentum)
• Learning Rate: 0.001 (with cosine annealing)
• Augmentation: RandAugment, Mosaic, Random Erasing
• Patience: 20 (early stopping)

Usage

from ultralytics import YOLO

# Load the model
model = YOLO('weights/best.pt')

# Run inference
results = model('path/to/image.jpg')

# Get predictions
for result in results:
    probs = result.probs
    print(f"Class: {result.names[probs.top1]}, Confidence: {probs.top1conf:.2f}")

Results

Training results and confusion matrices are available in the repository.

Dataset

The model was trained on a custom dataset of Roxy Migurdia images.

License

Apache 2.0