-
Notifications
You must be signed in to change notification settings - Fork 492
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Documentation of epsilon parameter #330
Conversation
Thanks for this Claudia; I'm very keen to get this merged and a release out that includes this, but I am going to be travelling for the next several days, so I may be a little slow in reviewing this. Please don't take my relative slowness as a lack on enthusiasm for this work. I will be getting to this as soon as I can. Thanks again! |
No problem at all, it's not urgent :) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for this -- only minor typo changes, which you can address easily.
Co-Authored-By: Leland McInnes <leland.mcinnes@gmail.com>
Co-Authored-By: Leland McInnes <leland.mcinnes@gmail.com>
Co-Authored-By: Leland McInnes <leland.mcinnes@gmail.com>
Thanks, I just commited all of your changes. |
Thank you! I'll merge this and see if I can get a release out soon. |
I have added a paragraph to the "Parameter Selection for HDBSCAN*" chapter and a demonstration of how epsilon can be used (same example as in the paper) in the new chapter "Combining HDBSCAN* with DBSCAN", which I added to index.rst.
Please feel free to correct possible grammar mistakes and to reword sentences that don't sound right to you, since English is not my first language.
By the way, I also have a Python script similar to examples/plot_hdbscan that shows how to use the epsilon parameter and allows to reproduce the GPS clusterings shown in the tutorial as well as all the other plots used in the paper. If you think that would be helpful then I could add this script to /examples, but I would then also need to add a folder "examples/data" with text files containing the data. Some of them come frome external sources (I'm using the jain, spiral and flame toy data sets
from http://cs.joensuu.fi/sipu/datasets/) and some are my own. I'm not sure where I should cite the origin of the data, maybe in a separate text file "_DataSource"?
Also, I was wondering whether it is possible to update HDBSCAN on conda-forge such that installing hdbscan with conda recognizes the new parameter?