[SPARK-11262] Unit test for gradient, loss layers, memory management for multilayer perceptron

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.5.1
• Fix Version/s: 2.0.0
• Component/s: ML, Tests
• Assignee: Alexander Ulanov
• Reporter: Alexander Ulanov
• Shepherd: Xiangrui Meng
• Created: 22/Oct/15 18:46
• Updated: 01/Apr/16 06:48
• Resolved: 01/Apr/16 06:48

Description

Multi-layer perceptron requires more rigorous tests and refactoring of layer interfaces to accommodate development of new features.
1)Implement unit test for gradient and loss
2)Refactor the internal layer interface to extract "loss function"

Issue Links

• is related to: SPARK-9471 Multilayer perceptron classifier
• is related to: SPARK-10478 Improve spark.ml.ann implementations for MLP
• links to: [Github] Pull Request #9229 (avulanov)