systemml-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From GitBox <...@apache.org>
Subject [GitHub] [systemml] j143-bot edited a comment on issue #613: [WIP][SYSTEMML-1748] Functionalize Kmeans
Date Wed, 13 Mar 2019 15:14:49 GMT
j143-bot edited a comment on issue #613: [WIP][SYSTEMML-1748] Functionalize Kmeans
URL: https://github.com/apache/systemml/pull/613#issuecomment-472462950
 
 
   Hi @opengjenkins - thanks for adding this example notebook.
   
   Your notebook is fine, have two nice examples. Also, your [dml script](https://github.com/apache/systemml/blob/b9f9cb85fd3807915939113d76b358a58319effa/scripts/staging/Kmeans-functional.dml)
in well into `11` nice funcs of which `4` main funcs.
   
   Although, this is good enough. Since the top level [K-means script](https://github.com/apache/systemml/blob/master/scripts/algorithms/Kmeans.dml)
is well tested, reusing most of it can scale your algorithm even better.
   
   PS: A functionalized K-means could help us lead the deep learning support path for systemml
    - one DL + kmean example is [Deep-K-Means](https://github.com/Sandbox3aster/Deep-K-Means-pytorch)
   i.e., applying k-means clustering on the weights, compression is achieved through weight
sharing and storing only the center of the clusters. Also, the compression ratio could be
as much as `40%`!

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services

Mime
View raw message