Ashish Sinha comments

Results 25 comments of


                                            Ashish Sinha

I haven't tried training the model yet.

when tested with FC100 dataset, it performs almost similar results as the miniImageNet results mentioned in the paper

yes

Thanks for the suggestion. I'll try it out and see if it works.

I take argmax of classes per pixel as suggested by @jiwoon-ahn . But I haven't yet run the code so can't say for sure if it would work.

My loss decreases to a very less extent, like from .68 to around .43 or .48. I use log_softmax with NLLLoss. What about you @xiangzhang06 ?

Ok i'll try

I would like to take a dig at this.

I think you would have to obtain a simplical complex first, and use LevelSetLayer to compute the PH.

following up on this! @amirhertz