Skip to content
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

Multivariate layout #14

Open
Make42 opened this issue Aug 7, 2019 · 3 comments
Open

Multivariate layout #14

Make42 opened this issue Aug 7, 2019 · 3 comments

Comments

@Make42
Copy link

Make42 commented Aug 7, 2019

As far as I understand the layout returns for each node in the graph a 2d-vector. I would like to provide the number of dimensionality of the vectors and get this as a result. In other words, I would like to layout my graph in N dimensional coordinate space. Obviously I will not be able to display the result on a 2D screen, but that would be okay, since I would like to use the result in further analysis steps. How can I adapt your algorithm to provide this feature?

@bhargavchippada
Copy link
Owner

@Make42 Sorry for the late reply, there is a 3D version of the algorithm and it must be an extension of the 2D layout algorithm. But I seem to remember reading in the paper somewhere that this algorithm might not scale to higher dimensions. I need to verify this claim again.

@Make42
Copy link
Author

Make42 commented May 2, 2020

@bhargavchippada Could you tell me which paper you are talking about and provide a link if possible?

@bhargavchippada
Copy link
Owner

bhargavchippada commented May 3, 2020

@Make42 I found that somebody implemented a ndim forceatlas algorithm here, not sure if it's valid to do this for > 3, you can check out how they did it. I created a milestone for this, especially to implement for 3D positions and extend it to n_dim.

I am unable to find the place where I read what I was talking about earlier. Probably, the idea is that this algorithm is built/tuned for 2D & 3D visualizations so extending it to higher dimensions may not give valid results which makes sense. you cannot visualize them but maybe use as features elsewhere to see if they are making sense, this is a nice experiment.
A similar request -> #3

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants