Krsto Proroković

### Describe the bug Should the code L273-L280 https://github.com/scikit-learn/scikit-learn/blob/364c77e047ca08a95862becf40a04fe9d4cd2c98/sklearn/mixture/_base.py#L273-L280 be indented under the for loop? https://github.com/scikit-learn/scikit-learn/blob/364c77e047ca08a95862becf40a04fe9d4cd2c98/sklearn/mixture/_base.py#L237 ### Steps/Code to Reproduce Have a look at the bug description. ### Expected Results...

Refactor Scholar.Neighbors

1

As argued in #254, we need to update `Scholar.Neighbors` in the following way: - [x] Add `BruteKNN` module that implements brute-force k-NN search. Its `predict` function has to return `{neighbor_indices,...

We already have PCA implemented. Two extensions that might be worth adding: - [x] Incremental PCA. This was already requested [here](https://elixirforum.com/t/how-to-execute-incremental-pca-or-similar-batch-tasks-in-nx-scholar/59039). - [ ] Kernel PCA.

[ENH] Mahalanobis distance

16

#### Reference Issues/PRs Fixes #1225. #### What does this implement/fix? Explain your changes. Implements Mahalanobis distance. #### Any other comments? Following ["Learning a Mahalanobis Distance based Dynamic Time Warping Measure...

I believe there is a reduction missing (probably `Nx.mean/2`) in the implementation of cross-entropy inside `Scholar.Linear.LogisticRegression`. https://github.com/elixir-nx/scholar/blob/975938a4e57704ad2400aa89ec18761cc0e45ce4/lib/scholar/linear/logistic_regression.ex#L188 `xs` is a tensor of shape `{num_samples, num_features}`, `ys` is a tensor of...