ManoharSai2000
ManoharSai2000
I have been using the same pretrained model and this is either a wrong model or an incomplete one. You can download this here. https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnet50_ra_224-8efdb4bb.pth
I have the same doubt! and my dataset size is only 240 images
Same Problem here??
I have carefully studied the Deconvolutional SSD paper where the details of SSD512 implementation are mentioned and replaced the backbone, prior Anchors correspondingly!
It can be viewed such that the similar labels follow similar distribution and hence their mix, and they do not change rapidly over training as in case of GAN's generator...