pytorch-image-models
pytorch-image-models copied to clipboard
[BUG] Inquiry into feature_only setting of ghostnet backbone
Hi. We successfully introduced the ghostnet backbone from timm.
When we set feature_only as true, the structures after block 7 are ignored. The backbone works well but we are wondering if this is normal? Why these layers are abandoned?
@yorklyb it's a bug, last feature info needs to include up to this line https://github.com/rwightman/pytorch-image-models/blob/master/timm/models/ghostnet.py#L174
I need to check more closely to see what the current state and fix it, but have other priorities on the go right now...
actually not really an issue / bug due to head order, fewer layers than described in the issue are chopped off and don't feel it makes sense to change