[FLINK-5841] Algorithms for each pipeline stage should handle NaN, infinity like in scikit-learn

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Won't Do
• Component/s: Library / Machine Learning
• Assignee: Stavros Kontopoulos
• Reporter: Stavros Kontopoulos
• Created: 18/Feb/17 17:54
• Updated: 28/Feb/19 22:57
• Resolved: 28/Feb/19 22:57

Description

Algorithms in scikit-learn don't accept NaN, Infinity values. Since we are following the scikit-learn approach we should conform to that.
Right now values are propagated... check pre-processing algos.