File size: 3,165 Bytes
89b1f42 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 |
import argparse
import subprocess
def run_transfer_learning(dataset, epochs, batch_size, imgsz, patience, cache, pretrained, cos_lr, profile, plots, resume, augment, model, run):
# Construct the command with user-provided arguments
command = [
"python", "./experiment/transfer_learning_train&test.py",
"--dataset", str(dataset),
"--epochs", str(epochs),
"--batch", str(batch_size),
"--imgsz", str(imgsz),
"--patience", str(patience),
"--cache", cache,
"--model", model,
"--run", run
]
# Append boolean flags conditionally
if pretrained:
command.append("--pretrained")
if cos_lr:
command.append("--cos_lr")
if profile:
command.append("--profile")
if plots:
command.append("--plots")
if resume:
command.append("--resume")
if augment:
command.append("--augment")
# Use subprocess to run the script with the arguments
subprocess.run(command, check=True)
if __name__ == "__main__":
parser = argparse.ArgumentParser(description="Run transfer learning with YOLO model.")
# Define all arguments with default values from your script
parser.add_argument('--dataset', type=str, choices=["Birds-Nest", "Common-VALID", "Electric-Substation", "InsPLAD-det"], help='Dataset name to be used')
parser.add_argument("--epochs", type=int, default=1000, help="Number of epochs")
parser.add_argument("--batch", type=int, default=16, help="Batch size")
parser.add_argument("--imgsz", type=int, default=640, help="Image size")
parser.add_argument("--patience", type=int, default=30, help="Patience for early stopping")
parser.add_argument("--cache", type=str, default='ram', help="Cache option")
parser.add_argument("--pretrained", action="store_true", help="Use pretrained weights")
parser.add_argument("--cos_lr", action="store_true", help="Use cosine learning rate")
parser.add_argument("--profile", action="store_true", help="Profile the training")
parser.add_argument("--plots", action="store_true", help="Generate plots")
parser.add_argument("--resume", action="store_true", help="Resume run")
parser.add_argument("--augment", action="store_true", help="Apply augmentation techniques during training")
parser.add_argument("--model", type=str, choices=["yolov8n", "yolov8s", "yolov8m", "yolov8l", "yolov10n", "yolov10s", "yolov10m", "yolov10l"], help="Model to use")
parser.add_argument("--run", type=str, choices=["From_Scratch", "Finetuning", "freeze_[P1-P3]", "freeze_Backbone", "freeze_[P1-23]"], help="Run mode")
args = parser.parse_args()
# Call the function to run transfer learning
run_transfer_learning(
dataset=args.dataset,
epochs=args.epochs,
batch_size=args.batch,
imgsz=args.imgsz,
patience=args.patience,
cache=args.cache,
pretrained=args.pretrained,
cos_lr=args.cos_lr,
profile=args.profile,
plots=args.plots,
resume=args.resume,
augment=args.augment,
model=args.model,
run=args.run
) |