Open AndriyMulyar opened 5 years ago
Mzazvor
commented
4 years ago Hi, I am currently doing research for my master thesis which is also related to Decision trees and i would love compare oblique trees to other used methods. I have managed to run your package but there just seems to be no way to visualize or at least get to learned tree model. Is there a easy way to do it ?
Best requards Martin Zázvoka
AndriyMulyar
commented
4 years ago No there is no current implementation. It can be implemented as the hyperplane weights are accessible through Cython code. Look at the predict function.
Thanks, Andriy
On Mon, Dec 30, 2019, 5:19 AM Mzazvor notifications@github.com wrote:
Hi, I am currently doing research for my master thesis which is also related to Decision trees and i would love compare oblique trees to other used methods. I have managed to run your package but there just seems to be no way to visualize or at least get to learned tree model. Is there a easy way to do it ?
Best requards Martin Zázvoka
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/AndriyMulyar/sklearn-oblique-tree/issues/2?email_source=notifications&email_token=ADJ4TBW5KM5E4NWURUPACZDQ3HDLNA5CNFSM4GYPWC7KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEHZ7FPY#issuecomment-569635519, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADJ4TBQMM6PMBNECIHKBSRDQ3HDLNANCNFSM4GYPWC7A .
Mzazvor
commented
4 years ago Thanx for quick reply.I will look into that. -------- Původní zpráva --------Od: Andriy Mulyar notifications@github.com Datum: 30.12.19 16:11 (GMT+01:00) Komu: AndriyMulyar/sklearn-oblique-tree sklearn-oblique-tree@noreply.github.com Cc: Mzazvor zazvorka.martin@gmail.com, Comment comment@noreply.github.com Předmět: Re: [AndriyMulyar/sklearn-oblique-tree] Add visualization. (#2) No there is no current implementation. It can be implemented as the
hyperplane weights are accessible through Cython code. Look at the predict
function.
Thanks,
Andriy
On Mon, Dec 30, 2019, 5:19 AM Mzazvor notifications@github.com wrote:
Hi,
I am currently doing research for my master thesis which is also related
to Decision trees and i would love compare oblique trees to other used
methods. I have managed to run your package but there just seems to be no
way to visualize or at least get to learned tree model. Is there a easy way
to do it ?
Best requards
Martin Zázvoka
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
or unsubscribe
https://github.com/notifications/unsubscribe-auth/ADJ4TBQMM6PMBNECIHKBSRDQ3HDLNANCNFSM4GYPWC7A
.
—You are receiving this because you commented.Reply to this email directly, view it on GitHub, or unsubscribe.
[
{
"@context": "http://schema.org",
"@type": "EmailMessage",
"potentialAction": {
"@type": "ViewAction",
"name": "View Issue"
},
"description": "View this Issue on GitHub",
"publisher": {
"@type": "Organization",
"name": "GitHub",
"url": "https://github.com"
}
}
]
Mzazvor
commented
4 years ago Hello again, After some time i was finally able to get into it and i just found out, that i would have to learn Cython, because i have no idea how to do it properly. It seemes that there is option to write learned tree into string in original C code, but the section is just commented out. Is this done on purpouse ( dt_file arg is in python api, but not used in C code)
Thanx Martin
AndriyMulyar
commented
4 years ago Hi Martin,
Yes I commented it out. The function did not work when running the C compiled C binary.
The original implementation contains an excellent visualization tool. Extend it into the python interface.