sramirez / spark-MDLP-discretization

Spark implementation of Fayyad's discretizer based on Minimum Description Length Principle (MDLP)

sramirez/spark-MDLP-discretization Issues

AbstractMethodError in Spark 2.3.0
Updated 6 years ago3
Add jenkinsfile to enable CI building in jenkins
Updated 6 years ago3
DiscretizerModelReader::load fails with Spark 2.1.1
Updated 7 years ago1
Something wrong with vector?
Closed 7 years ago1
few values threshold method does not seem to be deterministic
Updated 7 years ago46
Create 0.2 release a publish to https://spark-packages.org/ (or some other central place)
Closed 7 years ago4
Error when label is date
Closed 7 years ago4
Upgrade to spark 2.x
Closed 7 years ago13
Handling metadata
Closed 7 years ago
Possible performance issue on 40k row dataset
Closed 8 years ago9
Splits sometimes have excessive precision
Closed 8 years ago4
Sometimes buckets with too few instances are generated
Closed 8 years ago8
Include cache call in unit tests
Closed 8 years ago4
Need to handle nulls as a separate bin
Closed 8 years ago2
Match error for some datasets
Closed 8 years ago2
Null values in the label (target) cause NPE
Closed 8 years ago4
Need way to control split stopping criteria
Closed 8 years ago5
Why are the column(s) to bin required to be a vector?
Closed 8 years ago1
Publish on Spark Packages Main Repository
Closed 8 years ago6
maxBins not honored
Closed 8 years ago3
Add some unit tests
Closed 8 years ago4
Generated splits are missing leading -Infinity
Closed 8 years ago2
Update to scala 11 and spark 1.6.2
Closed 8 years ago1