CODEWITHSUNDEEP
pythonscikit-learn
Mohamed Ali JAMAOUI
Mohamed Ali JAMAOUI

Reputation: 14689

How to add an sklearn wrapper for a new ML algorithm

I would like to integrate factorization machines in sklearn. I checked sklearn documentation and the web for how to wrap a new algorithm but this requirement seems to be not very well documented.

So, I would like to ask on whether there is a documentation on how to add a new algorithm wrapper to sklearn (besides reading the source code)?

Upvotes: 7

Views: 4139

Answers (2)

inversion
inversion

Reputation: 1304

After working through the sklearn documentation, the best thing to do is to look through a complete working example.

The XGBoost module has a thorough sklearn wrapper, which you can see here:

https://github.com/dmlc/xgboost/blob/master/python-package/xgboost/sklearn.py

Upvotes: 6

Andrés Marafioti
Andrés Marafioti

Reputation: 952

From this FAQ I get they are not very fond of new algorithms http://scikit-learn.org/stable/faq.html#selectiveness for reasons that seem valid. Given that, it is plausible to think that there are not any documentation on how to add a new algorithm wrapper. I will add that I've been using the package for a while now and I've never found anything either on their website or other websites that was similar to what you're looking for.

Upvotes: 0

Related Questions