Github user coderxiang commented on the pull request:
https://github.com/apache/spark/pull/458#issuecomment-41002133
@yinxusen Just sent the data/code to you. For the running time, from the
web UI, it appears to be just (nIter * average reduce/aggregate), . The
aggregate time for SGD remains almost identical in each iteration, while ADMM's
reduce time varies and often taker longer time in later iterations. Since ADMM
has early termination criterion, it usually takes less than 500 iterations
before convergence. I may include these details in following evaluation.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
