You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@systemds.apache.org by GitBox <gi...@apache.org> on 2021/01/22 19:47:07 UTC

[GitHub] [systemds] erwinTU commented on pull request #1125: [WIP] Scikit-learn converter

erwinTU commented on pull request #1125:
URL: https://github.com/apache/systemds/pull/1125#issuecomment-765645135


   Hi. Sorry for the delay in answering.
   
   I think we looked into quite a different direction at first, but after reading your comments we have now implemented a very simple proof of concept. Especially the idea with the pipeline was helpful (@mboehm7).
   
   This POC uses a dumped (using pickle) Sklearn pipeline and produces a simple dml script:
   
   ```
   Sklearn pipeline:
   Pipeline(steps=[('standardscaler', StandardScaler()), ('kmeans', KMeans())])
   
   DML Script
   source("scripts/builtin/scale") as ns_scale
   source("scripts/builtin/kmeans") as ns_kmeans
   
   step_0 = ns_scale::m_scale(X, True, True)
   step_1 = ns_kmeans::m_kmeans(step_0, 8, 10, 300)
   ```
   
   Is this approach alright? 
   @Shafaq-Siddiqi 


----------------------------------------------------------------
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