|
|
--- |
|
|
language: |
|
|
- en |
|
|
base_model: |
|
|
- Ultralytics/YOLOv8 |
|
|
datasets: |
|
|
- jonalvarez01/reto2g2-dataset |
|
|
--- |
|
|
|
|
|
|
|
|
# Mobile and Hat Detection Model - YOLOv8 |
|
|
|
|
|
This repository contains a YOLOv8 model trained to detect mobile phones and hats in images. |
|
|
|
|
|
## Model Description |
|
|
|
|
|
The model was trained using YOLOv8n architecture to detect two classes: |
|
|
- Mobile phones |
|
|
- Hats |
|
|
|
|
|
### Training Details |
|
|
- Base model: YOLOv8n |
|
|
- Training epochs: 100 |
|
|
- Hardware: CUDA-enabled GPU |
|
|
- Framework: Ultralytics YOLO |
|
|
|
|
|
## Usage |
|
|
|
|
|
```python |
|
|
from ultralytics import YOLO |
|
|
|
|
|
# Load the model |
|
|
model = YOLO('path_to_model.pt') |
|
|
|
|
|
# Perform detection on an image |
|
|
results = model('path_to_image.jpg') |
|
|
``` |
|
|
|
|
|
## Training Code |
|
|
|
|
|
The model was trained using the following script: |
|
|
|
|
|
```python |
|
|
from ultralytics import YOLO |
|
|
|
|
|
# Load a model |
|
|
model = YOLO("yolov8n.pt").to('cuda') |
|
|
# Train the model |
|
|
results = model.train(data="data.yaml", epochs=100) |
|
|
``` |
|
|
|
|
|
## Dataset |
|
|
|
|
|
The model was trained on a custom dataset containing images of mobile phones and caps. The dataset was structured following YOLO format requirements. |
|
|
|
|
|
|
|
|
## Model Performance Metrics |
|
|
|
|
|
The model's performance was evaluated over 100 epochs of training. Here are the key metrics: |
|
|
|
|
|
<img src="metrics/results.png" width="600" /> |
|
|
|
|
|
### Confusion Matrix |
|
|
|
|
|
<img src="metrics/confusion_matrix.png" width="400" /> |
|
|
|
|
|
### Precision-Recall Curve |
|
|
|
|
|
|
|
|
<img src="metrics/P_curve.png" width="400" /> |
|
|
<img src="metrics/R_curve.png" width="400" /> |
|
|
|
|
|
### F1-Score Curve |
|
|
|
|
|
<img src="metrics/F1_curve.png" width="400" /> |
