-
Notifications
You must be signed in to change notification settings - Fork 418
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
Adding epoch cost functionality to both algorithms. #79
Conversation
Now the cost of the k-modes algorithm at each epoch can be accessed through the `epoch_costs_` attribute of a `KModes` object after fitting to some data. These are particularly useful when trying to visualise or analyse the convergence of the k-modes algorithm.
See previous commit.
Thanks, @daffidwilde ! This looks good, but could you have a look at the failing tests, please? |
@nicodv Sorry about how long it's been now. Everything seems to be fine on my fork but I hadn't seen the incorporation of parallelism (something else I had thought of opening a ticket about). Would you rather I resolved these conflicts locally or are you happy to edit yourself? |
Just had a proper look over the new implementation. I'll redo this now. |
Sorry for forgetting about this PR. Merged now. Thanks for the contribution, @daffidwilde ! FYI, I've added a simple test case for this too: 8d0e1fc |
Following issue #59, I have added the ability to access the epoch costs in the 'best' iteration of either algorithm. This is done by using the
epoch_costs_
attribute of aKModes
orKPrototypes
object.I am open to suggestions with regard to my writing style and I hope to hear from you soon.
Cheers