DeepLabV3+ Parameters
joshmyersdean opened this issue · comments
Hello!
Thank you for this nice library! I had a question about the implementation of DeepLabV3+. I believe the model with a R50 backbone should have ~42M params but when I count the parameters in your implementation it is closer to ~26M. What would cause this discrepancy?
def count_parameters(model):
return sum(p.numel() for p in model.parameters() if p.requires_grad)
net = DeepLabV3Plus(encoder_name="resnet50", classes = 21)
print(count_parameters(net)) # prints 26,682,725
This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this will be closed in 7 days.
This issue was closed because it has been stalled for 7 days with no activity.