lx-cly / YOLOv7_OBB

YOLOv7+KLD
GNU General Public License v3.0
35 stars 8 forks source link

How does your code detect an argument? "names" #18

Open Suppersine opened 1 year ago

Suppersine commented 1 year ago

I was running a custom training with this .ipynb command `# Commence Training P6-Models %cd ~/yolov7-obb-lxcly/

!python train_aux.py \ --weights yolov7-w6.pt \ --cfg cfg/training/yolov7-w6.yaml \ --data ~/datasets/roboflow/dataabs.yaml \ --hyp data/hyp.scratch.custom.p6.yaml \ --epochs 20 \ --batch-size 16 \ --img-size 1280 1280 \ --workers 4 \ --device 0 \ --exist-ok \ --name 'YOLOV7-test-p6'`

and then your train_aux.py throws this error

image

lx-cly commented 12 months ago

File"train_aux.py" ,Line245,in train, Add an argument "names" to the "create_dataloader" create_dataloader(train_path, imgsz, batch_size, gs, opt, names,...)

Suppersine commented 12 months ago

Also this one. what makes the dimensions of targets & gain unequal in your {--cfg yolov7-w6.yaml} subarchitecture when run by train_aux,py? messageImage_1696932652248