-
Notifications
You must be signed in to change notification settings - Fork 86
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Metric MDS #180
Metric MDS #180
Conversation
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.
Sorry I didn't notice your review request earlier! Belated review below. Thanks for adding this, it looks like it may be useful!
|
||
An arbitrary transformation function can be provided to `metric` parameter to | ||
perform metric MDS with transformed proximities. The function has to accept two parameters, | ||
a vector of proximities and a vector of distances, in order to calculate disparities |
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.
What do the vectors represent? I.e., if p
is a proximity vector, what is the meaning of p[i]
? Likewise, what is the meaning of d[i]
?
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.
Proximity and distance vectors are compact (1D) representations of upper triangular part of proximity and distance matrices correspondingly. I'll add more information to the docs about this.
|
||
Calculate a symmetric Euclidean (L2) distance matrix. | ||
""" | ||
L2distance(X::AbstractMatrix{T}) where {T<:Real} = L2distance!(zeros(T,size(X,2),size(X,2)), X) |
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.
Presumably we could use Matrix{T}(undef, size(X, 2), size(X, 3))
but the performance gain is unlikely to be observable in practice.
Implementation of weighted (non)metric MDS with transformed proximities.