Mahout reads data from the Hive Viewbuilder. Lets update the profile test so that it actually outputs real data.
Maybe that means playing with the constants and thresholds to make them as low as possible so that some kind of recommendations are made,
or conversly, maybe we need to increase the size of the test data set to 40 or 100 or 200 records?
FYI You have to "turn off" the hadoop-common dependency since the base mahout dependency from maven is a 1x versoin, in order to run develop using the mahout profile .
Mahout reads data from the Hive Viewbuilder. Lets update the profile test so that it actually outputs real data.
FYI You have to "turn off" the hadoop-common dependency since the base mahout dependency from maven is a 1x versoin, in order to run develop using the mahout profile .