Oleg Kachan
Oleg Kachan
It would be nice to have anisotropy in all directions, not only along the axes. For n-dimensional image it could be parameterized by an n x n matrix defining the...
It would be nice to have IDTxl installable using tools like `pip` and/or `conda` - [ ] pip - [ ] conda
As far as I understand this implementation of CCF do not have a capability to predict probability of a class. It would be nice to implement this.
## Describe the bug As my datapoints are empirical distributions I want to use the Wasserstein distance as a loss function over a batch of shape `(n_batch, n_points, dimension)`. Standard...
I use custom scorer from sklearn, via `make_scorer` function. It does not work if `needs_proba=True`, so metrics like ROC AUC, PR AUC are unfeasible to be used with `smote_variants`. The...
I'm using the [latest](https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes/rel-23-08.html) NVIDIA-supplied container based on Ubuntu 22.04, Python 3.10, G++ 11.4.0, CMake 3.22.1, and is unable to install GUDHI via pip, getting the following message ``` pip...
Dear Pedro, Can you explain lines 177, 180, 184 in utils.py? `p = np.sqrt(disp_target / disp_source)` `target_pow_train['covs'] = np.stack([powm(covi, 1.0/p) for covi in covs_target_train])` `target_pow_test['covs'] = np.stack([powm(covi, 1.0/p) for covi...
Implement differentiable computation of the Frechet mean for various manifolds
Based on the paper: Efficient Riemannian Optimization on the Stiefel Manifold via the Cayley Transform https://arxiv.org/abs/2002.01113
Good day, My question is whether Inform is still an active project. Are there will to consider the implementation of kNN-based estimators of entropy and mutual information, i.e. Kozachenko-Leonenko and...