[MAHOUT-2004] Properly route all CUDAMMul calls to the correct algorithms

• Type: New Feature
• Status: Resolved
• Priority: Major
• Resolution: Won't Fix
• Affects Version/s: 0.13.1
• Fix Version/s: 0.13.2
• Epic Link: Mahout-CUDA
• Assignee: Andrew Palumbo
• Reporter: Andrew Palumbo
• Created: 14/Jul/17 21:04
• Updated: 31/Jan/24 21:51
• Resolved: 31/Jan/24 21:51

Description

CUDAMMul has been implemented and is currently routing Sparse Sparse and Dense Dense mmul calls correctly, Sparse Dense must be properly routed.. (as well as Dense Sparse).