ã«ã¹ã¿ã ããŒã¿ãéãã
ãã®ã¬ã€ãã§ã¯ãðã§ç¬èªã®ã«ã¹ã¿ã ããŒã¿ã»ããããã¬ãŒãã³ã°ããæ¹æ³ã説æããŸãã YOLOv5ð.
å§ããåã«
ã¬ããã¯ããŒã³ããrequirements.txtãPython>=3.8.0ç°å¢ã«ã€ã³ã¹ããŒã«ããã PyTorch>=1.8.ã¢ãã«ãšããŒã¿ã»ããã¯ãææ°ã®YOLOv5 ãªãªãŒã¹ããèªåçã«ããŠã³ããŒããããŸãã
git clone https://github.com/ultralytics/yolov5 # clone
cd yolov5
pip install -r requirements.txt # install
ã«ã¹ã¿ã ããŒã¿ã§ãã¬ãŒãã³ã°
ãªããžã§ã¯ããæ€åºããã«ã¹ã¿ã ã¢ãã«ã®äœæã¯ãç»åã®åéãšæŽçãé¢å¿ã®ãããªããžã§ã¯ãã®ã©ãã«ä»ããã¢ãã«ã®ãã¬ãŒãã³ã°ãäºæž¬ãè¡ãããã®å®ç°å¢ãžã®å±éããããŠãã®å±éãããã¢ãã«ã䜿çšããŠãšããžã±ãŒã¹ã®äŸãåéããç¹°ãè¿ããšæ¹åãè¡ããšããå埩çãªããã»ã¹ã§ãã
ã©ã€ã»ã³ã¹
Ultralytics ã¯2ã€ã®ã©ã€ã»ã³ã¹ã»ãªãã·ã§ã³ãæäŸããŠããïŒ
- AGPL-3.0 ã©ã€ã»ã³ã¹ã¯ãåŠçãæ奜家ã«çæ³çãªOSIæ¿èªã®ãªãŒãã³ãœãŒã¹ã©ã€ã»ã³ã¹ã§ãã
- ãšã³ã¿ãŒãã©ã€ãºã»ã©ã€ã»ã³ã¹ã¯ãåœç€Ÿã®AIã¢ãã«ã補åããµãŒãã¹ã«çµã¿èŸŒãããšããäŒæ¥åãã®ã©ã€ã»ã³ã¹ã§ãã
詳ããã¯Ultralytics ã©ã€ã»ã³ã¹ãã芧ãã ããã
YOLOv5 ã¢ãã«ã¯ããã®ããŒã¿å ã®ãªããžã§ã¯ãã®ã¯ã©ã¹ãåŠç¿ããããã«ãã©ãã«ä»ããããããŒã¿ã§åŠç¿ãããªããã°ãªããªããåŠç¿ãéå§ããåã«ããŒã¿ã»ãããäœæããã«ã¯ã2ã€ã®ãªãã·ã§ã³ãããïŒ
ãªãã·ã§ã³1 RoboflowããŒã¿ã»ãã
1.1 ç»åã®åé
ããªãã®ã¢ãã«ã¯ããææ¬ãèŠãŠåŠã³ãŸããå®éã®æ®åœ±ã«è¿ãç»åã§ãã¬ãŒãã³ã°ããããšãæãéèŠã§ããçæ³çã«ã¯ãæçµçã«ãããžã§ã¯ããå±éããã®ãšåãæ§æïŒã«ã¡ã©ãã¢ã³ã°ã«ãç §æãªã©ïŒãããããŸããŸãªç»åãåéããããšã§ãã
ãããäžå¯èœãªå Žåã¯ãå ¬éããŒã¿ã»ããããå§ããŠåæã¢ãã«ãèšç·Žããæšè«äžã«éçã®ç»åããµã³ããªã³ã°ããŠããŒã¿ã»ãããšã¢ãã«ãå埩çã«æ¹åããããšãã§ããŸãã
1.2 ã©ãã«ã®äœæ
ç»åãåéããããã¢ãã«ãåŠç¿ããããã®ã°ã©ã³ããã¥ã«ãŒã¹ãäœæããããã«ãé¢å¿ã®ãããªããžã§ã¯ãã«æ³šéãä»ããå¿ èŠããããŸãã
Roboflow Annotateã¯ã·ã³ãã«ãªãŠã§ãããŒã¹ã®ããŒã«ã§ãããŒã ã§ç»åã管çããã©ãã«ãä»ããŠãYOLOv5'泚éãã©ãŒãããã§ãšã¯ã¹ããŒãããŸãã
1.3 ããŒã¿ã»ããã®æºåYOLOv5
ç»åã«Roboflow ã©ãã«ãä»ãããã©ããã«é¢ããããããŒã¿ã»ãããYOLO ãã©ãŒãããã«å€æããYOLOv5 YAML èšå®ãã¡ã€ã«ãäœæãããã¬ãŒãã³ã°ã¹ã¯ãªããã«ã€ã³ããŒãããããã«ãã¹ãããããã«äœ¿çšããããšãã§ããŸãã
Roboflow ç¡æã¢ã«ãŠã³ããäœæãã ã«ã¢ããããŒãããŸãã Public
ã¯ãŒã¯ã¹ããŒã¹ã§ã泚éã®ãªãç»åã«ã©ãã«ãä»ããããŒã¿ã»ãããçæã㊠YOLOv5 Pytorch
ãšãã圢åŒããšã£ãŠããã
泚ïŒYOLOv5 ã¯ãã¬ãŒãã³ã°äžã«ãªã³ã©ã€ã³ã§ãªãŒã°ã¡ã³ããŒã·ã§ã³ãè¡ããããYOLOv5 ã§ã®ãã¬ãŒãã³ã°ã«Roboflow ã®ãªãŒã°ã¡ã³ããŒã·ã§ã³ã»ã¹ããããé©çšããããšã¯æšå¥šããªãããããã以äžã®ååŠçã¹ããããé©çšããããšãæšå¥šããïŒ
- ãªãŒããªãªãšã³ã- ç»åããEXIFãªãªãšã³ããŒã·ã§ã³ãåãé€ããŸãã
- ãªãµã€ãºïŒã¹ãã¬ããïŒ- ã¢ãã«ã®æ£æ¹åœ¢ã®å ¥åãµã€ãºã«åãããŸãïŒYOLOv5 ã®ããã©ã«ã㯠640x640ïŒã
ããŒãžã§ã³ãçæããããšã§ãããŒã¿ã»ããã®ã¹ãããã·ã§ãããåŸãããã®ã§ãåŸã§ç»åãè¿œå ããããèšå®ãå€æŽãããããŠãããã€ã§ããã®ããŒã¿ã»ããã«ããã®ãŒã£ãŠãä»åŸã®ã¢ãã«ã®ãã¬ãŒãã³ã°å®è¡ãæ¯èŒããããšãã§ããŸãã
èŒžåº YOLOv5 Pytorch
圢åŒã®ã¹ããããããã¬ãŒãã³ã°ã¹ã¯ãªãããŸãã¯ããŒãããã¯ã«ã³ããŒããŠãããŒã¿ã»ãããããŠã³ããŒãããŸãã
ãªãã·ã§ã³2ïŒæåããŒã¿ã»ããã®äœæ
2.1 äœæ dataset.yaml
COCO128 ã®æåã®128æã®ç»åãããªãå°èŠæš¡ãªãã¥ãŒããªã¢ã«ã»ããŒã¿ã»ããã®äŸã§ããã COCO train2017ããããã®åã128æã®ç»åãèšç·Žãšæ€èšŒã®äž¡æ¹ã«äœ¿çšããæã
ã®èšç·Žãã€ãã©ã€ã³ã以äžã®èœåãæã€ããšãæ€èšŒããã ãªãŒããŒãã£ããã£ã³ã°. data/coco128.yaml以äžã«ç€ºãã®ã¯ïŒ1) ããŒã¿ã»ããã®ã«ãŒãã»ãã£ã¬ã¯ããªãå®çŸ©ããããŒã¿ã»ããèšå®ãã¡ã€ã«ã§ããïŒ path
ãžã®çžå¯Ÿãã¹ãš train
/ val
/ test
ã€ã¡ãŒãžã»ãã£ã¬ã¯ã㪠*.txt
ãã¡ã€ã«ã®ç»åãã¹) ãšã2) ã¯ã©ã¹ names
ãšããèŸæžãããïŒ
# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
path: ../datasets/coco128 # dataset root dir
train: images/train2017 # train images (relative to 'path') 128 images
val: images/train2017 # val images (relative to 'path') 128 images
test: # test images (optional)
# Classes (80 COCO classes)
names:
0: person
1: bicycle
2: car
# ...
77: teddy bear
78: hair drier
79: toothbrush
2.2 ã©ãã«ã®äœæ
泚éããŒã«ã䜿ã£ãŠç»åã«ã©ãã«ãä»ããããã©ãã«ã次ã®ããã«ãšã¯ã¹ããŒãããŸãã YOLO ãã©ãŒãããäžäºº *.txt
ãã¡ã€ã«ïŒç»åã«ãªããžã§ã¯ãããªãå Žå㯠*.txt
ãã¡ã€ã«ãå¿
èŠïŒããã® *.txt
ãã¡ã€ã«ã®ä»æ§ã¯ä»¥äžã®éãïŒ
- ãªããžã§ã¯ãããšã«1è¡
- åè¡ã¯
class x_center y_center width height
ãšãã圢åŒããšã£ãŠããã - ããã¯ã¹åº§æšã¯ æ£èŠåxywh ãã©ãŒãããïŒ0ãã1ãŸã§ïŒãããã¯ã¹ããã¯ã»ã«ã®å Žåã¯
x_center
ãããŠwidth
ç»åã®å¹ ã§y_center
ãããŠheight
ç»åã®é«ãã«ãã - ã¯ã©ã¹çªå·ã¯ãŒãã€ã³ããã¯ã¹ïŒ0ããå§ãŸãïŒã
äžã®ç»åã«å¯Ÿå¿ããã©ãã«ãã¡ã€ã«ã«ã¯ã2人ã®äººç©ïŒã¯ã©ã¹ 0
ãšåŒãåãïŒã¯ã©ã¹ 27
):
2.3 ãã£ã¬ã¯ããªã®æŽç
YOLOv5 以äžã®äŸã«åŸã£ãŠãåè»ãšãã«ã®ç»åãšã©ãã«ãæŽçããŠãã ããã /coco128
ã®äžã«ããã /datasets
ãã£ã¬ã¯ã㪠暪 ãã® /yolov5
ãã£ã¬ã¯ããªã«ããã YOLOv5 åç»åã®ã©ãã«ãèªåçã«æ€çŽ¢ ã®æåŸã®ã€ã³ã¹ã¿ã³ã¹ã /images/
ã䜿çšããŠåç»åãã¹ã« /labels/
.äŸãã°ïŒ
3.ã¢ãã«ã®éžæ
åŠç¿ãéå§ããäºååŠç¿æžã¿ã¢ãã«ãéžæããŸããããã§ã¯ãå©çšå¯èœãªã¢ãã«ã®äžã§2çªç®ã«å°ãããæãé«éãªYOLOv5sãéžæããŸãããã¹ãŠã®ã¢ãã«ã®å®å šãªæ¯èŒã¯READMEã®è¡šãåç §ããŠãã ããã
4.é»è»
COCO128ã®YOLOv5sã¢ãã«ããããŒã¿ã»ãããããããµã€ãºãç»åãµã€ãºãæå®ããäºååŠç¿ããã --weights yolov5s.pt
(æšå¥šïŒããŸãã¯ã©ã³ãã ã«åæåããã --weights '' --cfg yolov5s.yaml
(æšå¥šããªãïŒãäºåèšç·Žãããéã¿ã¯ ææ°YOLOv5 ãªãªãŒã¹.
ããã
ð¡ è¿œå --cache ram
ãŸã㯠--cache disk
ãã¬ãŒãã³ã°ã®ã¹ããŒãã¢ããã®ããïŒããªãã®RAM/ãã£ã¹ã¯ãªãœãŒã¹ãå¿
èŠïŒã
ããã
ð¡ åžžã«ããŒã«ã«ããŒã¿ã»ãããããã¬ãŒãã³ã°ããŠãã ãããGoogle ãã©ã€ãã®ãããªããŠã³ãããããã©ã€ãããããã¯ãŒã¯ãã©ã€ãã¯éåžžã«é ãã
ãã¹ãŠã®ãã¬ãŒãã³ã°çµæ㯠runs/train/
ã©ã³ã»ãã£ã¬ã¯ããªãã€ã³ã¯ãªã¡ã³ããããã runs/train/exp2
, runs/train/exp3
ãªã©ã詳现ã¯ãã¥ãŒããªã¢ã«ããŒãã®ãã¬ãŒãã³ã°ã»ã¯ã·ã§ã³ãã芧ãã ããã
5.èŠèŠåãã
Comet ãã®ã³ã°ãšããžã¥ã¢ã©ã€ãŒãŒã·ã§ã³ ð NEW
Cometã¯ãYOLOv5 ãšå®å šã«çµ±åãããŸãããã¢ãã«ã®ã¡ããªã¯ã¹ããªã¢ã«ã¿ã€ã ã§è¿œè·¡ããã³å¯èŠåãããã€ããŒãã©ã¡ãŒã¿ãããŒã¿ã»ãããããã³ã¢ãã«ã®ãã§ãã¯ãã€ã³ããä¿åããComet ã«ã¹ã¿ã ããã«ã§ã¢ãã«ã®äºæž¬ãå¯èŠåããŸããComet ã¯ãäœæ¥å 容ãèŠå€±ãããšãªããããããèŠæš¡ã®ããŒã éã§çµæãå ±æããã³ã©ãã¬ãŒã·ã§ã³ããããšã容æã«ããŸãïŒ
å§ããã®ã¯ç°¡åã ïŒ
pip install comet_ml # 1. install
export COMET_API_KEY=<Your API Key> # 2. paste API key
python train.py --img 640 --epochs 3 --data coco128.yaml --weights yolov5s.pt # 3. train
ãã®çµ±åã§ãµããŒãããããã¹ãŠã®Comet æ©èœã®è©³çŽ°ã«ã€ããŠã¯ã以äžãã芧ãã ããã Comet ãã¥ãŒããªã¢ã«.Comet ã«ã€ããŠãã£ãšãç¥ãã«ãªãããæ¹ã¯ããã¡ããžã©ããã ããã¥ã¡ã³ããŒã·ã§ã³.ãŸãã¯ãComet ã³ã©ãã»ããŒãããã¯ããè©Šããã ããïŒ
ClearML ãã®ã³ã°ãšèªåå ð NEW
ClearMLã¯YOLOv5 ã«å®å šã«çµ±åãããŠãããå®éšã®è¿œè·¡ãããŒã¿ã»ããã®ããŒãžã§ã³ç®¡çãããã«ã¯ãªã¢ãŒãã§ã®ãã¬ãŒãã³ã°å®è¡ãå¯èœã§ããClearML ãæå¹ã«ããã«ã¯ã以äžã®ãªã³ã¯ãã¯ãªãã¯ããŠãã ããïŒ
pip install clearml
- èµ°ã
clearml-init
ClearML ãµãŒããŒã«æ¥ç¶ãã
ã©ã€ãã¢ããããŒããã¢ãã«ã®ã¢ããããŒããå®éšã®æ¯èŒãªã©ãå®éšãããŒãžã£ãŒããæåŸ ãããçŽ æŽãããæ©èœã¯ãã¹ãŠåŸãããŸãããClearML ãäŸãã°ã³ããããããŠããªãå€æŽãã€ã³ã¹ããŒã«ãããããã±ãŒãžã远跡ããŸãããã®ãããã§ãClearML ã¿ã¹ã¯ïŒç§ãã¡ã¯ãããå®éšãšåŒãã§ããïŒã¯ç°ãªããã·ã³ã§ãåçŸå¯èœã§ããïŒãã£ã1è¡è¿œå ããã ãã§ãYOLOv5 ãã¬ãŒãã³ã°ã¿ã¹ã¯ããã¥ãŒã«ã¹ã±ãžã¥ãŒã«ããä»»æã®æ°ã®ClearML ãšãŒãžã§ã³ãïŒã¯ãŒã«ãŒïŒãå®è¡ã§ããã
ClearML Dataã䜿ã£ãŠããŒã¿ã»ãããããŒãžã§ã³ç®¡çããäžæã®IDã䜿ã£ãŠYOLOv5 ãããããããšã§ãäœèšãªæéãå¢ããããšãªãããŒã¿ã远跡ããããšãã§ããã詳ããã¯ClearML ãã¥ãŒããªã¢ã«ãã芧ãã ããïŒ
ããŒã«ã«ã»ãã®ã³ã°
ãã¬ãŒãã³ã°çµæ㯠ãã³ãœã«ããŒã ãã㊠ã·ãŒãšã¹ã〠ãã¬ãŒã runs/train
ãšããŠãæ°ãããã¬ãŒãã³ã°ããšã«æ°ããå®éšãã£ã¬ã¯ããªãäœæãããã runs/train/exp2
, runs/train/exp3
çã
ã
ãã®ãã£ã¬ã¯ããªã«ã¯ãèšç·Žãšvalã®çµ±èšãã¢ã¶ã€ã¯ãã©ãã«ãäºæž¬ãæ¡åŒµã¢ã¶ã€ã¯ã®ã»ãã粟床- åçŸçïŒPRïŒæ²ç·ãæ··åè¡åãªã©ã®ã¡ããªã¯ã¹ãã°ã©ããå«ãŸããŠããŸãã
çµæãã¡ã€ã« results.csv
ãæŽæ°ããããã³ã« ãšããã¯ãšããŠããããããã results.png
(äžå³)ã¯ãã¬ãŒãã³ã°çµäºåŸã«è¡šç€ºãããŸãããŸã results.csv
ãã¡ã€ã«ãæåã§äœæããïŒ
from utils.plots import plot_results
plot_results("path/to/results.csv") # plot 'results.csv' as 'results.png'
次ã®ã¹ããã
ã¢ãã«ããã¬ãŒãã³ã°ãããããæé©ãªãã§ãã¯ãã€ã³ãã䜿ãããšãã§ããã best.pt
ã«ïŒ
- å®è¡ãã CLIãŸã㯠Pythonæ°ããç»åããããªã«å¯Ÿããæšè«
- ãã¬ã€ã³ããã«ãããã¹ãã®åå²ã§ç²ŸåºŠã æ€èšŒããã
- ãšã¯ã¹ããŒãå TensorFlowKerasãONNX ãTFliteãTF.jsãCoreML ããã³TensorRT ãã©ãŒããã
- ãã€ããŒãã©ã¡ãŒã¿ãé²åãããŠããã©ãŒãã³ã¹ãåäžããã
- å®äžçã®ç»åããµã³ããªã³ã°ããŠããŒã¿ã»ããã«è¿œå ããããšã§ãã¢ãã«ãæ¹åããŸãã
察å¿ç°å¢
Ultralytics ãã¯ãããšããå¿ èŠäžå¯æ¬ ãªäŸåé¢ä¿ãããªã€ã³ã¹ããŒã«ããããããŸããŸãªããã«äœ¿ããç°å¢ãæäŸããã CUDAãCUDNNã Pythonãã㊠PyTorchãªã©ãããªã€ã³ã¹ããŒã«ãããŠããŸãã
- ç¡æGPU ããŒã:
- Google ã¯ã©ãŠã GCPã¯ã€ãã¯ã¹ã¿ãŒãã¬ã€ã
- ã¢ããŸã³ AWSã¯ã€ãã¯ã¹ã¿ãŒãã¬ã€ã
- AzureïŒAzureML ã¯ã€ãã¯ã¹ã¿ãŒãã¬ã€ã
- ããã«ãŒ: Dockerã¯ã€ãã¯ã¹ã¿ãŒãã¬ã€ã
ãããžã§ã¯ãç¶æ³
ãã®ãããžã¯ãYOLOv5 GitHub ActionsContinuous Integration (CI) ãã¹ãããã¹ãŠæ£åžžã«ãã¹ããŠããããšã瀺ããŸãããããã®CIãã¹ãã¯ããã¬ãŒãã³ã°ãæ€èšŒãæšè«ããšã¯ã¹ããŒãããã³ãããŒã¯ãšãã£ãããŸããŸãªéèŠãªåŽé¢ã«ããã£ãŠãYOLOv5 ã®æ©èœãšããã©ãŒãã³ã¹ãå³å¯ã«ãã§ãã¯ããŸãããããã®ãã¹ãã¯ãmacOSãWindowsãUbuntuäžã§ã24æéããšããã³æ°ããã³ãããããšã«å®æœãããäžè²«ããä¿¡é Œæ§ã®é«ãåäœãä¿èšŒããŸãã
ããããã質å
ã«ã¹ã¿ã ããŒã¿ã»ããã§YOLOv5 ããã¬ãŒãã³ã°ããã«ã¯ïŒ
ã«ã¹ã¿ã ããŒã¿ã»ããã§YOLOv5 ããã¬ãŒãã³ã°ããã«ã¯ãããã€ãã®ã¹ããããããïŒ
- ããŒã¿ã»ãããæºåããç»åãåéããã©ãã«ãä»ãããã®ãããªããŒã«ã䜿ã£ãŠ Roboflowãªã©ã®ããŒã«ã䜿ã£ãŠããŒã¿ãæŽçããYOLOv5 圢åŒã§ãšã¯ã¹ããŒãããã
- ã»ããã¢ããç°å¢:YOLOv5 ãªããžããªãã¯ããŒã³ããäŸåé¢ä¿ãã€ã³ã¹ããŒã«ããïŒ
- ããŒã¿ã»ããæ§æã®äœæ:ãæžã
dataset.yaml
ãã¡ã€ã«ã§train/valã®ãã¹ãšã¯ã©ã¹åãå®çŸ©ããã - ã¢ãã«ãèšç·Žãã:
YOLOv5 ããŒã¿ã»ããã«æ³šéãä»ããã«ã¯ãã©ã®ãããªããŒã«ã䜿ãã°ããã§ããïŒ
Roboflow Annotateã¯ãç»åã«ã©ãã«ãä»ããããã®çŽæçãªãŠã§ãããŒã¹ã®ããŒã«ã§ããããŒã ã³ã©ãã¬ãŒã·ã§ã³ããµããŒãããYOLOv5 圢åŒã§ãšã¯ã¹ããŒãããŸããç»åãåéããããRoboflow ã䜿çšããŠã泚éãå¹ççã«äœæããã³ç®¡çããŸãããã®ä»ã®ãªãã·ã§ã³ãšããŠãLabelImgãCVATã®ãããªããŒã«ã«ã¢ãããŒã·ã§ã³çšã®ããŒã«ããããŸãã
YOLO ã¢ãã«ã®ãã¬ãŒãã³ã°ã«Ultralytics HUB ã䜿ãã¹ãçç±ã¯ïŒ
Ultralytics HUBã¯ããã¬ãŒãã³ã°ããããã€ãYOLO ã¢ãã«ç®¡çã®ããã®ãšã³ãããŒãšã³ãã®ãã©ãããã©ãŒã ãæäŸããŸããUltralytics HUBã䜿çšããå©ç¹ã¯ä»¥äžã®éãã§ãïŒ
- ç°¡åãªã¢ãã«ãã¬ãŒãã³ã°ïŒäºåã«èšå®ãããç°å¢ã§ãã¬ãŒãã³ã°ããã»ã¹ãç°¡çŽ åããŸãã
- ããŒã¿ç®¡çïŒããŒã¿ã»ãããšããŒãžã§ã³ç®¡çã容æã«ããŸãã
- ãªã¢ã«ã¿ã€ã ã¢ãã¿ãªã³ã°ïŒä»¥äžã®ãããªããŒã«ãçµ±åã Cometãªã©ã®ããŒã«ãçµ±åãããªã¢ã«ã¿ã€ã ã®ã¡ããªã¯ã¹è¿œè·¡ãšå¯èŠåãå®çŸããŸãã
- ã³ã©ãã¬ãŒã·ã§ã³ïŒãªãœãŒã¹ãå ±æããç°¡åã«ç®¡çã§ããããŒã ãããžã§ã¯ãã«æé©ã§ãã
泚éä»ãããŒã¿ãYOLOv5 圢åŒã«å€æããã«ã¯ïŒ
泚éä»ãããŒã¿ãRoboflow ã䜿ã£ãŠYOLOv5 圢åŒã«å€æããïŒ
- ããŒã¿ã»ããã Roboflow ã¯ãŒã¯ã¹ããŒã¹ã«ã¢ããããŒãããŸãã
- ãŸã ã©ãã«ã貌ãããŠããªãå Žåã¯ãç»åã«ã©ãã«ã貌ãã
- çæãšãšã¯ã¹ããŒã ã®ããŒã¿ã»ãã
YOLOv5 Pytorch
ãã©ãŒãããã䜿çšãããæ£æ¹åœ¢ã®å ¥åãµã€ãºïŒäŸïŒ640x640ïŒã«èªåãªãªãšã³ãããªãµã€ãºïŒã¹ãã¬ããïŒãªã©ã®ååŠçã¹ããããé©çšãããŠããããšã確èªããã - ããŒã¿ã»ãããããŠã³ããŒãããŠãYOLOv5 ãã¬ãŒãã³ã°ã¹ã¯ãªããã«çµ±åããŸãã
åçšã¢ããªã±ãŒã·ã§ã³ã§YOLOv5 ã䜿çšããå Žåã®ã©ã€ã»ã³ã¹ãªãã·ã§ã³ã¯ïŒ
Ultralytics ã¯2ã€ã®ã©ã€ã»ã³ã¹ã»ãªãã·ã§ã³ãæäŸããŠããïŒ
- AGPL-3.0 ã©ã€ã»ã³ã¹éåçšã«é©ãããªãŒãã³ãœãŒã¹ã©ã€ã»ã³ã¹ã§ãåŠçãæ奜家ã«æé©ã§ãã
- ãšã³ã¿ãŒãã©ã€ãºã»ã©ã€ã»ã³ã¹ïŒYOLOv5 ãåçšè£œåããµãŒãã¹ã«çµ±åããããšããäŒæ¥åãã§ãã詳现ã«ã€ããŠã¯ãã©ã€ã»ã³ã¹ã®ããŒãžãã芧ãã ããã
詳ããã¯ãUltralytics ã©ã€ã»ã³ã¹ã«é¢ããã¬ã€ããã芧ãã ããã