Nathan
Nathan
Thank you for the detailed response. That gives me a good place to start, if I can get around to it, hopefully!
> Hi @nmichlo, > > Thanks for reporting this. I have a couple questions for you that will help us understand this better: > > 1. On which type of...
@brimoor `random_split` would add items to different splits if more data is added over time, even if a seed is specified? Maybe a `deterministic_split` could be added? Where a hash...
That makes sense. - Maybe they don't have to be full blown benchmarks but partial ones just to get an idea of the behaviour and runtimes of the different algorithms?...
(Admittedly this is from legacy code that I have not written myself, so I am not entirely sure about the original intention) But what I can see is that we...
Hi there @IceClear, thank you for the kind words! Unfortunately it is not possible to calculate these metrics directly on MNIST, the dataset itself is not intended for disentanglement. Disentanglement...
I am closing this issue for now. If it is still a problem, please don't hesitate to re-open this issue!
Hi @meffmadd, really glad you are finding it useful! I managed to get away with wandb results a while back so I never got around to fixing this. I'll investigate...
You noted the object is an OmegaConf instance. I don't think it would break too much if we switch that over to a dictionary, and recursively convert all the values....
Thank you so much for the PR! I left a few comments about tests. We just need to make sure to add the new keys to the configs and (possibly)...