YOLO-NAS 예제

https://github.com/Deci-AI/super-gradients

GitHub - Deci-AI/super-gradients: Easily train or fine-tune SOTA computer vision models with one open source training library. T

요구사항

• 파이썬 버전 : 3.7, 3.8, 3.9
• PyTorch 버전 : 1.9.0 <= torch < 1.14
• Nvidia CUDA Toolkit >= 11.2
• CuDNN >= 8.1.x
• Nvidia Driver with CUDA >= 11.2 support (≥460.x)

YOLO-NAS 설치

pip install super-gradients

예제 코드

from super_gradients.training import models

model = models.get('yolo_nas_l', pretrained_weights='coco').cuda()
input_path = 'vid/people.mp4'
output_path = 'output/output.mp4'
model.predict(input_path, conf=0.25).save(output_path)

# 사전 학습된 모델 불러오기 및 GPU 사용 설정
from super_gradients.training import models
model = models.get('yolo_nas_l', pretrained_weights='coco').cuda()

# 감지할 영상 파일 설정 및 출력 경로 설정 후 저장
input_path = 'vid/people.mp4'
output_path = 'output/output.mp4'
model.predict(input_path, conf=0.25).save(output_path)

결과물