search

bethatkinson / rpart

Recursive Partitioning and Regression Trees
43 stars 23 forks source link

Interface to python #49

Open nickgiki opened 1 year ago

nickgiki commented 1 year ago

Is there any way to call rpart through python (without using rpy2 or R altogether)? If not, would anyone be interested in investing some time to create an rpart-python package with me?

bethatkinson commented 1 year ago

I'm sure there must be implementations of decision trees in python, but I'm not familiar with what has been done. I don't use python so I can't help you out. Sorry. Beth

From: nickgiki @.> Sent: Thursday, June 1, 2023 4:54 AM To: bethatkinson/rpart @.> Cc: Subscribed @.***> Subject: [EXTERNAL] [bethatkinson/rpart] Interface to python (Issue #49)

Is there any way to call rpart through python (without using rpy2 or R altogether)? If not, would anyone be interested in investing some time to create an rpart-python package with me?

— Reply to this email directly, view it on GitHubhttps://github.com/bethatkinson/rpart/issues/49, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ACWQG56M7KONB75I3YRE7KDXJBRFFANCNFSM6AAAAAAYWVI224. You are receiving this because you are subscribed to this thread.Message ID: @.***>

nickgiki commented 1 year ago

The truth is that rpart, is much superior to all of the Decision Tree implementations in python. It produces better performing trees, without the need for preprocessing and also handles missing values. That's why exposing it to python would be an instant success. I could try it on my own but I'm missing is the C part. If anyone you know would be interested in investing some time in this, I'm glad to contribute.