[HAMA-822] Add feature transformer interface to improve the power and flexibility of existing machine learning model

• Type: New Feature
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.7.0
• Fix Version/s: 0.7.0
• Component/s: machine learning
• Labels: machine_learning
• Assignee: Yexi Jiang
• Reporter: Yexi Jiang
• Created: 26/Nov/13 04:52
• Updated: 09/Dec/13 03:33
• Resolved: 09/Dec/13 03:33

Description

Add the feature transformer interface to improve the power of the existing learning model.

If the user implicitly know how to transform and extract the features, the user can implement the concrete feature transformer to extract the useful information from the original features.

The implemented feature transformer can reduce the dimension of the original features to improve the generalization of the model. Or it can transform the original nonlinearly separable features by mapping them to a linear separable space.