[SPARK-19110] DistributedLDAModel returns different logPrior for original and loaded model

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.3.1, 1.4.1, 1.5.2, 1.6.3, 2.0.2, 2.1.0, 2.2.0
• Fix Version/s: 2.0.3, 2.1.1, 2.2.0
• Component/s: ML, MLlib
• Target Version/s: 2.0.3, 2.1.1, 2.2.0
• Assignee: Miao Wang
• Reporter: Miao Wang
• Shepherd: Joseph K. Bradley
• Created: 06/Jan/17 20:38
• Updated: 10/Jan/17 00:14
• Resolved: 07/Jan/17 19:42

Description

While adding DistributedLDAModel training summary for SparkR, I found that the logPrior for original and loaded model is different.
For example, in the test("read/write DistributedLDAModel"), I add the test:
val logPrior = model.asInstanceOf[DistributedLDAModel].logPrior
val logPrior2 = model2.asInstanceOf[DistributedLDAModel].logPrior
assert(logPrior === logPrior2)
The test fails:
-4.394180878889078 did not equal -4.294290536919573

Issue Links

• links to: [Github] Pull Request #16491 (wangmiao1981)
• links to: [Github] Pull Request #16524 (wangmiao1981)