ARG-NCTU/rtdetr-spscd_smd13clsbuoy_finetuned_e50

RT-DETR Training Results

Model Information

• Model: runs/rtdetr/spscd_base_1280_b8_e100/weights/best.pt
• Dataset: finetune_dataset.yaml
• Classes: 13
• Image Size: 1280×1280
• Batch Size: 8
• Epochs: 50
• Workers: 16

Training Configuration

• Initial Learning Rate: 1e-05
• Final Learning Rate Factor: 0.01
• Patience: 50
• Mixed Precision (AMP): Enabled

Training Time

• Start: 2026-01-12 00:01:08
• End: 2026-01-12 08:37:43
• Duration: 8h 36m 35s (30,995 seconds)

Final Test Results

• mAP50: 0.1615
• mAP50-95: 0.1314

Files in this Repository

• best.pt - Best model checkpoint
• results.csv - Complete training metrics
• results.png - Training curves visualization
• confusion_matrix.png - Confusion matrix
• F1_curve.png, PR_curve.png - Performance curves
• samples/ - Sample validation predictions
• gpu_stats.csv - GPU monitoring data (utilization, memory, temperature, power) recorded per epoch

Note: TensorBoard logs are included if available. You can also visualize training using results.csv and the curve images.

How to Use

from ultralytics import RTDETR

# Load model
model = RTDETR('best.pt')

# Run inference
results = model('your_image.jpg')

Visualizing Training Progress

• View training curves in results.png
• Analyze per-class performance in confusion matrices
• Check results.csv for detailed epoch-by-epoch metrics