TransmogrifAI
TransmogrifAI copied to clipboard
Null Imputation for DateUnitCircleVectorizer
Related issues Issue
Describe the proposed solution Using (1, 0) instead of (0, 0) for null default value.
Describe alternatives you've considered
Alternatives do not only concern this transformer but the other vectorizer
that can return the mode as imputation technique.
Instead of getting the mode, randomly select an existing non null value so that the distribution of the feature is not changed.
However, this remains difficult :
- DateToUnitCircleTransformer is not an estimator
- As an estimator, you would store as a fitted param all the distinct non null values of the dataset.
Additional context
This is in the context where we have this HourOfDay circular representation of a MM-DD-YYYY 00h00m00s
date not being thrown out by SanityChecker because of Variance not being 0.
Thanks for the contribution! It looks like @mweilsalesforce is an internal user so signing the CLA is not required. However, we need to confirm this.
Codecov Report
Merging #555 (49425a8) into master (8dc772a) will decrease coverage by
0.00%
. The diff coverage is100.00%
.
:exclamation: Current head 49425a8 differs from pull request most recent head 0cab7fb. Consider uploading reports for the commit 0cab7fb to get more accurate results
@@ Coverage Diff @@
## master #555 +/- ##
==========================================
- Coverage 86.78% 86.78% -0.01%
==========================================
Files 347 347
Lines 12026 12024 -2
Branches 403 387 -16
==========================================
- Hits 10437 10435 -2
Misses 1589 1589
Impacted Files | Coverage Δ | |
---|---|---|
...s/impl/feature/DateMapToUnitCircleVectorizer.scala | 100.00% <100.00%> (ø) |
|
...ges/impl/feature/DateToUnitCircleTransformer.scala | 100.00% <100.00%> (ø) |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update f82a301...0cab7fb. Read the comment docs.
How do I get rid of this CLA stuff?
@michaelweilsalesforce did you fill out the form ?
Yes.
kicking cla bot
@michaelweilsalesforce do you have your salesforce email associated with @mweilsalesforce as well ? Or did you accidentally switch accounts and not realize it ? The cla bot is seeing two authors and it needs both authors to sign cla.
@michaelweilsalesforce do you have your salesforce email associated with @mweilsalesforce as well ? Or did you accidentally switch accounts and not realize it ? The cla bot is seeing two authors and it needs both authors to sign cla.
Both emails are [email protected] and [email protected]
Somehow I couldn't attach my salesforce account because it was linked to github enterprise. That's why I have created this other account