Closed jnwang closed 8 years ago
Random split may have some issues. https://github.com/sjyk/sampleclean-async/blob/master/src/main/scala/sampleclean/activeml/ActiveLearningAlgorithm.scala#L111
Please test it for record-level deduplication with the input argument of ActiveLearningParameters, budget = 60, batchSize = 10, and bootstrapSize = 10.
Random split may have some issues. https://github.com/sjyk/sampleclean-async/blob/master/src/main/scala/sampleclean/activeml/ActiveLearningAlgorithm.scala#L111
Please test it for record-level deduplication with the input argument of ActiveLearningParameters, budget = 60, batchSize = 10, and bootstrapSize = 10.