rapidsai/cuml

[FEA] Add c++ util for singleToDoublePrecisionPerfRatio

Open

#803 opened on Jun 26, 2019

View on GitHub
 (0 comments) (0 reactions) (0 assignees)Python (644 forks)auto 404
? - Needs TriageCUDA / C++Cython / Pythonfeature requestgood first issue

Repository metrics

Stars
 (5,223 stars)
PR merge metrics
 (PR metrics pending)

Description

This is needed so that we can warn users in the Python layer when they are using a device that might have lowered performance in double precision computations.

This ticket entails adding a function to ml_cuda_utils.h which calls singleToDoublePrecisionPerfRatio in the cuda runtime API and wrapping this utility function in Cython.

Contributor guide

Research direction
Implement the singleToDoublePrecisionPerfRatio call in a C++ header file under ml cuda utils.h, then create a Cython wrapper to expose it to Python. Follow existing patterns in the codebase for Cython integration.
Tech stack
cpppython
Domain
backendmachine learning
Issue type
Feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
Half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
C++CUDACython
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
50