graphtools

graphtools copied to clipboard

Hello, For a while now igraph has been using name `igraph` instead of `python-igraph` on PyPI. Could you please update the name you use when specifying dependencies? I believe the...

szhorvat

I'm wondering if the output of `G = graphtools.Graph(X,decay=None)` is working as intended. My impression was that with `decay=None`, a kNN graph would be created. However, inspection of `G.K` reveals...

stanleyjs

Hello, I have used graphtools to perform MNN batch correction in PHATE and it works well. I was wondering if I can use this same approach for other dimensionality techniques...

jluthy

Good day, This PR began its life to fix the issues with randomized svd discussed in #58. To fix that, I monkey patched in a version of randomized svd using...

stanleyjs

Hi, This test seems to fail in a way that is dependent on the sampling of the test dataset. I have been tinkering with this for a little while trying...

stanleyjs

Hi, I noticed that master fails `test_landmark.test_landmark_knn_graph` and `test_landmark.test_landmark_knn_pygsp_graph`. The reason these tests fails is due to the shape of the landmark operator. The tests expect the landmark transitions operator...

stanleyjs

logger.debug -> logger.log_debug

2

dburkhardt

Inaccuracy in randomized SVD at moderate ranks

7

Hi, ### *Randomized SVD is not accurate* Currently most (if not all) of the PCA / linear dimensionality reduction / SVD is first routed through either `TruncatedSVD` or `PCA(svd_solver='randomized')`. It...

stanleyjs

Need to rename theta (symmetrization) and beta (batch correction weight)

dburkhardt

graphtools should provide certain canonical manifold learning datasets. MATLAB implementation courtesy of Nick Marshall ``` function [X,C,name] = benchmark(code,N,alpha) % BENCHMARK Generates manifold learning benchmark datasets. % [X,C,name] = benchmark(code,N,alpha)...

scottgigante