|
|
--- |
|
|
title: YOLO Model |
|
|
tags: |
|
|
- yolo |
|
|
- object-detection |
|
|
- computer-vision |
|
|
- unknown |
|
|
- aegis-ai |
|
|
library_name: ultralytics |
|
|
license: agpl-3.0 |
|
|
--- |
|
|
|
|
|
# YOLO Model |
|
|
|
|
|
This model has been converted and optimized using the **Aegis AI Model Conversion Tool**. |
|
|
|
|
|
## Model Details |
|
|
|
|
|
- **Original Model**: Unknown |
|
|
- **Format**: UNKNOWN |
|
|
- **Task**: Object Detection |
|
|
- **Framework**: Ultralytics YOLO |
|
|
- **License**: AGPL-3.0 |
|
|
|
|
|
## Performance Metrics |
|
|
|
|
|
| Metric | Value | |
|
|
|--------|--------| |
|
|
| Average FPS | N/A | |
|
|
| Inference Time | N/A ms | |
|
|
| Memory Usage | N/A MB | |
|
|
| Target Hardware | cpu | |
|
|
|
|
|
## Hardware Information |
|
|
|
|
|
- **Platform**: Unknown |
|
|
- **Device**: cpu |
|
|
- **Optimization**: Hardware-specific optimizations applied |
|
|
|
|
|
## Usage |
|
|
|
|
|
### Loading the Model |
|
|
|
|
|
```python |
|
|
# For ONNX models |
|
|
import onnxruntime as ort |
|
|
session = ort.InferenceSession("model.onnx") |
|
|
|
|
|
# For PyTorch models |
|
|
from ultralytics import YOLO |
|
|
model = YOLO("model.pt") |
|
|
|
|
|
# For TensorRT models (NVIDIA GPU) |
|
|
# Requires TensorRT runtime |
|
|
model = YOLO("model.engine") |
|
|
``` |
|
|
|
|
|
### Inference |
|
|
|
|
|
```python |
|
|
import numpy as np |
|
|
from PIL import Image |
|
|
|
|
|
# Load your image |
|
|
image = Image.open("path/to/image.jpg") |
|
|
|
|
|
# Run inference |
|
|
results = model(image) |
|
|
|
|
|
# Process results |
|
|
for result in results: |
|
|
boxes = result.boxes # Bounding boxes |
|
|
classes = result.names # Class names |
|
|
``` |
|
|
|
|
|
## Conversion Details |
|
|
|
|
|
This model was converted using the Aegis AI Model Conversion Tool with the following configuration: |
|
|
|
|
|
- **Precision**: fp32 |
|
|
- **Optimization Level**: standard |
|
|
- **Hardware Target**: cpu |
|
|
- **Conversion Date**: 2025-08-18 15:11:05 |
|
|
|
|
|
## Model Architecture |
|
|
|
|
|
Based on the YOLO (You Only Look Once) architecture, this model provides real-time object detection capabilities with optimized performance for the target hardware. |
|
|
|
|
|
### Input |
|
|
|
|
|
- **Shape**: 640x640 |
|
|
- **Format**: RGB images |
|
|
- **Normalization**: [0-1] range |
|
|
|
|
|
### Output |
|
|
|
|
|
- **Bounding Boxes**: Object locations |
|
|
- **Confidence Scores**: Detection confidence |
|
|
- **Class Predictions**: Object categories |
|
|
|
|
|
## Benchmarking |
|
|
|
|
|
The model has been benchmarked on the target hardware with the following results: |
|
|
|
|
|
```json |
|
|
{} |
|
|
``` |
|
|
|
|
|
## Hardware Compatibility |
|
|
|
|
|
This model has been optimized for: |
|
|
- **Primary**: cpu |
|
|
- **Platform**: Unknown |
|
|
|
|
|
For other hardware configurations, consider using the Aegis AI Model Conversion Tool to create optimized versions. |
|
|
|
|
|
## Citation |
|
|
|
|
|
If you use this model in your research or project, please cite: |
|
|
|
|
|
```bibtex |
|
|
@misc{aegis-ai-converted-model, |
|
|
title={Aegis AI Converted YOLO Model}, |
|
|
author={Aegis AI Team}, |
|
|
year={2025}, |
|
|
howpublished={\url{https://github.com/aegis-ai/model-conversion-tool}} |
|
|
} |
|
|
``` |
|
|
|
|
|
## Related Models |
|
|
|
|
|
- [Original YOLO Models](https://github.com/ultralytics/ultralytics) |
|
|
- [Aegis AI Model Zoo](https://huggingface.co/aegis-ai) |
|
|
|
|
|
## Support |
|
|
|
|
|
For issues with this converted model or the conversion tool: |
|
|
- [GitHub Issues](https://github.com/aegis-ai/model-conversion-tool/issues) |
|
|
- [Aegis AI Documentation](https://docs.aegis-ai.com) |
|
|
|
|
|
--- |
|
|
|
|
|
*This model was automatically converted and uploaded by the Aegis AI Model Conversion Tool.* |
|
|
|
