Github user takuti commented on the issue:
https://github.com/apache/incubator-hivemall/pull/66
Or, aggregating over `docid` and output single row as an struct is possible:
| docid | probabilities |
|:---:|:---|
|1| { 0:0.9957867647115234, 1:0.004213235288476648 } |
|2| { 0:0.0014898943734896843, 1:0.9985101056265103 } |
Here, hyper-parameters used in the E step as follows:
- number of topics
- alpha (prior)
- delta (check convergence)
Fortunately, by counting unique `label` values, UDAF can know the number of
topics when the `terminate` operation is launched. So, one possible direction
is to create new `OnlineLDAModel` in UDAF's `terminate` w/ the number of
topics, and we set default values to `alpha` and `delta` (or manually pass to
UDAF as options).
---
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.
---
