Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Not serializable exception during computeFeatureStats #362

Closed
alkersan opened this issue Jul 11, 2019 · 1 comment · Fixed by #363
Closed

Not serializable exception during computeFeatureStats #362

alkersan opened this issue Jul 11, 2019 · 1 comment · Fixed by #363

Comments

@alkersan
Copy link

Description
Serializability issue during one of the stages of model training:

... 
(TID 3067704) had a not serializable result: scala.collection.immutable.MapLike$$anon$2
Serialization stack:
    - object not serializable (class: scala.collection.immutable.MapLike$$anon$2)
...

Stacktrace leads to com.salesforce.op.filters.PreparedFeatures#summaries method, which is called from com.salesforce.op.filters.RawFeatureFilter#computeFeatureStats. The usage
of mapValues leads to a non serializable Map, which is a long standing scala bug.

I've attempted to fix that just by adding a .map(identity), which resolved the issue, i.e.:

predictors.mapValues(Summary(_)).map(identity)

Not sure about "correct" way of dealing with that

Additional context
Transmogrify version: 0.5.1 & 0.5.2. Though it seems to be applicable to current master, as the related code didn't change since then.

@tovbinm
Copy link
Collaborator

tovbinm commented Jul 11, 2019

Thanks for reporting this. Fixed in #363

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants