scala - Difference between org.apache.spark.ml.classification and org.apache.spark.mllib.classification -
i'm writing spark application , use algorithms in mllib. in api doc found 2 different classes same algorithm.
for example, there 1 logisticregression in org.apache.spark.ml.classification logisticregressionwithsgd in org.apache.spark.mllib.classification.
the difference can find 1 in org.apache.spark.ml inherited estimator , able used in cross validation. quite confused placed in different packages. there know reason it? thanks!
it's jira ticket
and design doc:
mllib covers basic selection of machine learning algorithms, e.g., logistic regression, decision trees, alternating least squares, , k-means. current set of apis contains several design flaws prevent moving forward address practical machine learning pipelines, make mllib scalable project.
the new set of apis live under
org.apache.spark.ml, ,o.a.s.mllibdeprecated once migrate featureso.a.s.ml.
Comments
Post a Comment