Sambit Panda

update MGC to use scipy as reference implementation in tests

2

#### Reference issue #### Type of change #### What does this implement/fix? #### Additional information

My issue is about ... #### Reproducing code example: ``` Sample code to reproduce the problem ``` #### Error message ``` Traceback (most recent call last): File "", line 1,...

incorporate block permutation into K-sample tests

35

**Is your feature request related to a problem? Please describe.** **Describe the solution you'd like** **Describe alternatives you've considered** **Additional context (e.g. screenshots)**

Add two-sample with kernel Fisher discriminant analysis

1

https://papers.nips.cc/paper/3335-testing-for-homogeneity-with-kernel-fisher-discriminant-analysis.pdf

paper: https://arxiv.org/abs/1710.09859 python code in here: https://github.com/neurodata/non-parametric-clustering/ it may be in here too: https://github.com/mariarizzo/energy

**Is your feature request related to a problem? Please describe.** **Describe the solution you'd like** **Describe alternatives you've considered** **Additional context (e.g. screenshots)**

MGC/Dcorr embeddings

1

there are papers about "distance correlation analysis" where they solve a CCA like problem, but using Dcorr like ideas. same for HSIC. http://proceedings.mlr.press/v28/chang13.pdf http://proceedings.mlr.press/v54/cowley17a.html

Test whether different types

1

**Is your feature request related to a problem? Please describe.** Currently, methods in `hyppo` cast the input values to a `float`. While this is fine in most cases, it would...