Aun3223/e-waste-battery-anomaly-detection-yolov8n

YOLOv8n Battery Anomaly Detector

This is a YOLOv8n (nano) model trained for detecting normal batteries and swollen batteries in e-waste images.

Model Description

• Architecture: YOLOv8n
• Task: Object Detection
• Classes: normal_battery, swollen_battery
• Training Data: Custom dataset of simulated e-waste images.

Training Details

• Run Name: battery_detector_v1-3
• Epochs: 50
• Image Size: 640x640
• Batch Size: 32

Performance Metrics (on validation set)

• mAP50: N/A
• mAP50-95: N/A

How to Use

This model can be loaded and used with the ultralytics library in Python.

from ultralytics import YOLO

# Load the trained model
model = YOLO('Aun3223/e-waste-battery-anomaly-detection-yolov8n') # or specify local path if downloaded

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

# Show results
for r in results:
    print(r.boxes) # Print bounding box detections
    r.show()       # Display the image with detections

Training Artifacts

This repository also contains additional training artifacts:

• best.pt: The best performing model weights.
• args.yaml: The full configuration and arguments used for training.
• results.csv: Detailed performance metrics per epoch.
• confusion_matrix.png: Visualization of the model's classification performance.
• results.png: Training plots showing loss, accuracy, etc.