[systemds] 02/02: [SYSTEMDS-3237] GMM parse seeding the Kmeans init

[ba...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

baunsgaard pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/systemds.git

commit ed6460a7520bfc3a19d7c335e6a323f5c121813d
Author: baunsgaard <ba...@tugraz.at>
AuthorDate: Fri Dec 3 17:37:56 2021 +0100

    [SYSTEMDS-3237] GMM parse seeding the Kmeans init
    
    The GMM builtin did not parse the seed to the Kmeans initialization.
    This commit fixes this.
    
    Closes #1472
---
 scripts/builtin/gmm.dml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/builtin/gmm.dml b/scripts/builtin/gmm.dml
index a28dfd9..a4530c4 100644
--- a/scripts/builtin/gmm.dml
+++ b/scripts/builtin/gmm.dml
@@ -114,7 +114,7 @@ return (Matrix[Double] weight, Matrix[Double] mean, Matrix[Double] sigma, Matrix
   resp = matrix(0, nrow(X), n_components)
   if(init_params == "kmeans") {
     [C, Y] = kmeans(X=X, k=n_components, runs=10,
-      eps=tol, is_verbose=FALSE, avg_sample_size_per_centroid=100)
+      eps=tol, is_verbose=FALSE, avg_sample_size_per_centroid=100, seed=seed)
     resp = ((resp + t(seq(1, n_components))) == Y)
   }
   else if(init_params == "random") {