@faustomorales What an amazingly promising module - thank you so so much!
I would now like to use it for font classification (distinguishing between a small number N of distinct fonts, for which I have sample alphabet images).
Do you think this would be feasible with the existing model architecture (and hence codebase), or would it require complete remodelling?
Is it possible to get a document-level goodness-of-fit / classification probability out of the NN? If so then I think I would be home and dry.
What about the font classification as a nuisance parameter? Could I train a neural network on top of the existing one?
@faustomorales What an amazingly promising module - thank you so so much!
I would now like to use it for font classification (distinguishing between a small number N of distinct fonts, for which I have sample alphabet images).
Do you think this would be feasible with the existing model architecture (and hence codebase), or would it require complete remodelling?
Is it possible to get a document-level goodness-of-fit / classification probability out of the NN? If so then I think I would be home and dry.
What about the font classification as a nuisance parameter? Could I train a neural network on top of the existing one?
Many thanks in advance for any and all help!