|
--- |
|
license: apache-2.0 |
|
datasets: |
|
- visdrone |
|
model-index: |
|
- name: ENOT-AutoDL/yolov8s_visdrone |
|
results: |
|
- task: |
|
type: object-detection |
|
dataset: |
|
type: visdrone |
|
name: visdrone |
|
metrics: |
|
- type: precision |
|
value: 49,4 |
|
name: mAP50(baseline) |
|
- type: precision |
|
value: 48,4 |
|
name: mAP50(GMACs x2) |
|
- type: precision |
|
value: 46,0 |
|
name: mAP50(GMACs x3) |
|
library_name: ultralytics |
|
pipeline_tag: object-detection |
|
tags: |
|
- yolov8 |
|
- ENOT-AutoDL |
|
- yolo |
|
- vision |
|
- ultralytics |
|
- object-detection |
|
--- |
|
# ENOT-AutoDL YOLOv8 optimization on VisDrone dataset |
|
|
|
This repository contains models accelerated with [ENOT-AutoDL](https://pypi.org/project/enot-autodl/) framework. |
|
We trained yolov8s on VisDrone dataset and used it as our baseline. |
|
Also we provide simple python script to measure flops and metrics. |
|
|
|
## YOLOv8 Small |
|
|
|
| Model | GMACs | Image Size | mAP50 | mAP50-95 | |
|
|---------------------------|:-----------:|:-----------:|:-----------:|:-----------:| |
|
| **[YOLOv8 Ultralytics Baseline](https://docs.ultralytics.com/datasets/detect/visdrone/#dataset-yaml)** | 14,28 | 640 | 40,2 | 24,2 | |
|
| **YOLOv8n Enot Baseline** | 8,57 | 928 | 42,9 | 26,0 | |
|
| **YOLOv8s Enot Baseline** | 30,03 | 928 | 49,4 | 30,6 | |
|
| **YOLOv8s (x2)** | 15,01 (x2) | 928 | 48,3 (-1,1) | 29,8 (-0,8) | |
|
| **YOLOv8s (x3)** | 10,01 (x3) | 928 | 46,0 (-3,4) | 28,3 (-2,3) | |
|
|
|
# Validation |
|
|
|
To validate results, follow this steps: |
|
|
|
1. Install all required packages: |
|
```bash |
|
pip install -r requrements.txt |
|
``` |
|
|
|
2. Use validation script: |
|
```bash |
|
python validate.py enot_neural_architecture_selection_x2/weights/best.pt --imgsz 928 |
|
``` |
|
|
|
3. Use measure_macs script: |
|
```bash |
|
python measure_macs.py enot_neural_architecture_selection_x2/weights/best.pt --imgsz 928 |
|
``` |
|
|