error evaluating the pruned model

[melika-sce]

Hi, thanks for this good repo, I wonder how can I load the pruned model with YOLOv8?
when I try to run the below command to get the validation results and GFLOP and number of layers, I get this error:
command:
yolo task=detect mode=val batch=1 model=YOLO-Pruning-RKNN/runs/detect/yolov8m_Prune50%/weights/best.pt imgsz=1024 data=VisDrone.yaml save_json=True

error:
return F.conv2d(input, weight, bias, self.stride, RuntimeError: Given groups=1, weight of size [24, 24, 3, 3], expected input[1, 12, 256, 256] to have 24 channels, but got 12 channels instead

[melika-sce]

well, it was solved by making a val.py file with below codes:

from ultralytics import YOLO
model = YOLO('prune_model.pt')
metrics = model.val()

but I still don't know how to get FLOPs of pruned model, any help would be appreciated!

[heyongxin233]

Thank you for your interest in my repo！You can refer to the calculate.py file to get the model Macs.