Alex Loftus

LaTeX not exporting to PDF properly

3

# Current behavior LaTeX math works fine within boostnote, but then upon exporting to a pdf all the LaTeX gets broken. [within boostnote](https://www.dropbox.com/s/2tj6ou4e6c027dp/Screen%20Shot%202020-09-19%20at%205.39.52%20PM.png?dl=0) [exported version](https://github.com/loftusa/graspy-personal/blob/master/notes/Out-of-sample-Extension-tang2013.pdf) # Expected behavior the PDF...

ValueError: Input 0 of layer "generator" is incompatible with the layer

1

Hi, I'm getting this error when I try to predict after training: ``` ValueError: Input 0 of layer "generator" is incompatible with the layer: expected shape=(None, 40, 40, 3), found...

Heatmap updates

15

- adds section in heatmap tutorial for hierarchy labels - adds ticklabels functionality in heatmap function that seaborn has - small bug fix and error checking changes

I don't see one anywhere e.g., this has autoGMM and k-means: https://microsoft.github.io/graspologic/latest/tutorials/index.html#clustering but no hierarchical

Ensure clean python 3.10.0 build path

2

I'm having issues installing graspologic on 3.10.0 right now (see fun screenshot). This may or may not be local to my computer. In any case, we should ensure we have...

Number of embedding dimensions for vertex nomination

2

Right now, `SpectralVertexNomination` doesn't have a way to set embedding dimension. It currently just uses whatever the default is when `n_components=None`, which I think is just the profile likelihood method....

https://projecteuclid.org/journals/annals-of-applied-statistics/volume-9/issue-3/Vertex-nomination-schemes-for-membership-prediction/10.1214/15-AOAS834.full Right now we have spectral nomination implemented for single graphs (good at large scales, according to the above paper) There are two other main schemes the paper discusses, canonical...

Give screeplot an `ax` argument

2

function[ here.](https://github.com/microsoft/graspologic/blob/dev/graspologic/plot/plot.py#L1129) Currently no way to specify an `ax`, most other plotting functions have that unless there's a good reason not to. I don't see a good reason for this...

[BUG] graspy.neurodata.io no longer exists

5

graspologic.neurodata.io doesn't work either

Make SVD code faster

10

I've found that [scipy's implementation of the lanczos algorithm](https://docs.scipy.org/doc/scipy/reference/generated/scipy.sparse.linalg.eigsh.html) is extremely fast for finding eigenvalues and eigenvectors of a big, sparse real symmetric square matrix (I'm pretty sure more so...