Greg Hamerly
Greg Hamerly
When I use pygraphviz to visualize something like the following: ``` graph G { foo [label="x"]; bar [label=""]; foo -- bar; } ``` I expect to see *no* label for...
Sort-means is giving slightly different results in some of my testing, compared to all the other algorithms. I have a test case where every other algorithm converges in 45 iterations,...
Suggestions from @cur33 . These can (and should) be broken into separate issues as is appropriate, and handled separately. ``` Proposed changes to core library: - Change Dataset::print to accept...
The Dataset class uses a 1-dimensional array which is indexed like a 2-dimensional array almost everywhere. It would be good to switch to an explicit 2-dimensional indexing scheme. This could...
Right now the code for random initialization in https://github.com/ghamerly/fast-kmeans/blob/84a0e151e1006e510a6ec95d6606404ee54127e9/src/general_functions.cpp uses `rand()`, which is initialized by `srand()` here: https://github.com/ghamerly/fast-kmeans/blob/84a0e151e1006e510a6ec95d6606404ee54127e9/src/driver-experiment.cpp We should move away from `rand()` and `srand()` towards C++ `` classes...
It would be really nice to have a test framework for all the basic functions that we have in this repository. There are already some assignment verification routines which are...
Migrated from https://github.com/BaylorCS/baylorml/issues/2 (@BenjaminHorn) I have a fairly big dataset (100m \* 10) , and as i calculated it would take around 8 hours to initialise the centers with init_centers_kmeanspp_v2....
Migrated from https://github.com/BaylorCS/baylorml/issues/3 (@BenjaminHorn) I guess this loop's https://github.com/ghamerly/baylorml/blob/master/fast_kmeans/general_functions.cpp#L217 function is to prevent a point to be chosen more then once as center. Let's say we have 10 points in...
This should be the code that was originally in https://github.com/BaylorCS/baylorml/pull/6 However, I have moved the patches and rebased them on the current master.