Jochen Görtler
Jochen Görtler
Yeah, that would be cool as it would also document the required steps.
Yes, I meant _rank_ in terms of linear algebra. I agree that it could cause confusion with `dimensionality`. I think a detailed description in the documentation could take care of...
Ok cool. I can take a look into that topic and hopefully come up with `rank` implementation.
Should we think about writing an efficient matrix implementation? It feels like there is no real way around it?
An efficient implementation for `core.matrix` is also not a big priority for me at the moment. As for the `rank`, I still think we should add it to interface, similar...
Looking at the [Pytorch implementation](https://pytorch.org/docs/stable/_modules/torch/nn/functional.html#dropout2d) of `torch.nn.functional.dropout2d` (which is used by `torch.nn.Dropout2d`), we should be able to just call `tch::Tensor::feature_dropout` without generating the bindings.
I'll work with `feature_dropout` a bit. > If you think it would be convenient to have some dropout2d/dropout3d functions in the nn module so that it's easier to discover these,...
There are some machine learning approaches where the covariance matrix is computed directly from the data. PCA comes first to my mind. However, in GPs the covariance matrix serves a...
Sorry for replying so late, for some reason this issue has slipped through. The inputs to the kernel function can be high-dimensional vectors; its result will be a scalar value....