Summary: 44 instances, 41 unique

Text	Count
//TODO make nulls pass through	1
// TODO: This should be based on the retrieved score value kind	1
val dvv = new DoubleVectorVector() //TODO try re-using	1
//TODO: Check if not using squared euclidean distance is ok	1
// TODO: Handle DateType, TimestampType and DecimalType for label	1
PACKAGE: "flaky"           #TODO fix flaky test so isolation is not needed	1
//TODO: is there a better way to convert to unsigned byte?	1
# TODO figure out why we cannot just grab from SparkContext on databricks	1
// TODO (for v2): support regression scenarios	1
//case _: Param[String] => StringInfo //TODO fix erasure issues	1
// TODO: We call cache in order to avoid a bug with catalyst where CMS seems to get stuck in a loop	3
// TODO add default values	1
//TODO figure out why build machines don't have proper codecs	1
* TODO: expose as user configurable parameter	1
val numLevels: Int = levels.length //TODO: add the maximum possible number of levels?	1
// TODO allow for dynamic allocation	1
val fvv = new FloatVectorVector() //TODO try re-using	1
case _: SocketException => Array() //TODO investigate why sockets fail sometimes	1
// Update the schema - TODO: create method that would generate GUID and add it to the scored model	2
//TODO figure out why this doesent work	1
// TODO: figure out more elegant way	1
// TODO: investigate using destroy()	1
// TODO make this check more robust	1
// TODO: do we have to worry about locale?	1
//TODO: StructField("Mean", DoubleType, true),	1
//TODO: add tests to ensure	1
// TODO do this by hooking deeper into spark,	1
// TODO: surface prediction confidence	1
//TODO: StructField("Mean Deviation", DoubleType, true),	1
//TODO: StructField("Mode", StringType, true))	1
// TODO (for v2): Introduce a referenceDistribution function param for user to override the uniform distribution	1
// TODO: list types	1
//TODO use this elsewhere	1
PACKAGE: "lightgbm.split1" #TODO speed up LGBM Tests and remove split	1
// TODO: evaluate all models; for now, get first model name found	1
//TODO: StructField("Range", DoubleType, true),	1
// TODO fix this spelling in 1.15 update	1
// TODO: The cache call is a workaround for issue 1075:	1
// TODO refactor to remove duplicate from TextAnalyticsBase	1
//TODO make this less brittle	1
*           TODO: Consider adding check for correct class name.	1