julien-c/timm-dpn92

PyTorch implementation of DualPathNetworks is based on original MXNet implementation . The training images in the dataset are usually photos clearly representing one of the 1,000 labels . Training images are mostly from the US (45.4%) and Great Britain (7.6%) As such the model will work better on images picturing scenes from these countries . The model will probably not generalize well on drawings or images containing multiple objects with different labels, to name a few, and it will not be generalized well . It is compatible with the pretrained weights from cypw's MXNet, which are based on a dataset of 14 millions of hand-annotated images with 1,,