Training hyperfast from scratch

[Shreyas-Bhat]

Thank you so much to the authors for open sourcing this code and model weights! they are of great use to the community. I was wondering if the authors have any pointers on training networks from scratch for a new dataset. That would be great.

[davidbp]

In order to reproduce the results and investigate further, may you share a training script to get to the pretrained weights that are loaded in the classifier?

[Shreyas-Bhat]

Hi @davidbp , sorry, I am unsure if this question is directed to me or to the authors. could you please clarify? Thanks!

[davidbp]

to the authors!