Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/7705#discussion_r35794294
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/clustering/LDAModel.scala ---
@@ -197,8 +238,86 @@ class LocalLDAModel private[clustering] (
// TODO:
// override def topicDistributions(documents: RDD[(Long, Vector)]):
RDD[(Long, Vector)] = ???
+ /**
+ * Calculate and return per-word likelihood bound, using the `batch` of
+ * documents as evaluation corpus.
+ */
+ // TODO: calcualte logPerplexity over training set online during
training, reusing gammad instead
+ // of performing variational inference again in [[bound()]]
+ def logPerplexity(
+ batch: RDD[(Long, Vector)],
+ totalDocs: Long): Double = {
--- End diff --
By average vs. total, I meant:
* average: perplexity normalized by number of docs in the sample. E.g., if
you double the number of docs by creating an extra copy of each one, then the
perplexity should stay the same.
* total: perplexity without normalization (as it is now). E.g., if you
double the number of docs by creating an extra copy of each one, then the
perplexity should double.
I prefer average since users can undo the normalization easily.
---
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.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
