-
Currently, in ComplementNB, we estimate class prior but do not use it. I think we can provide an option to consider class prior in ComplementNB (like other naive bayes algorithms in scikit-learn). Rea…
-
I'm thinking it might be worthwhile to try to replace the euclidean distances used in kd-tree and ball-tree here:
https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/neighbors/_dist_metr…
-
#### Describe the workflow you want to enable
We are proposing to add a cythonized module that allows for building oblique trees.
Oblique trees originally proposed by, Breiman 2001 [1], would enab…
-
Some analysis of our source code use suggests neither `SpectralBiclustering` nor `SpectralCoclustering` are known to be in public use. They have design issues that have not been addressed in many year…
-
#### Describe the workflow you want to enable
According to https://scikit-learn.org/stable/modules/clustering.html#silhouette-coefficient
> The Silhouette Coefficient is generally higher for conve…
-
#### Description
ValueError: "array must not contain infs or NaNs" thrown when fitting an ExpSineSquared kernel to large dimension data
#### Steps/Code to Reproduce
```py
import skle…
-
Poisson regression is very commonly used for [survival analysis](https://en.wikipedia.org/wiki/Survival_analysis). In this context, it is necessary to include the exposure time as a [log-offset](https…
-
Thank you for your sharing.
However, when in the true development environments, there are always having missing value.
It seems no missing value process for each algorithm.
-
### Describe the workflow you want to enable
The sparse linear regression problem is the NP-hard problem of performing ordinary L² linear regression with the catch that at most a fixed number of th…
-
I know and tested that 'ranger' is the faster R package. But after reading http://datascience.la/benchmarking-random-forest-implementations/ I have this question.
Maybe some benchmarks show that ran…