rapidsai/cuml

[FEA] Add c++ util for singleToDoublePrecisionPerfRatio

Open

#803 aberto em 26 de jun. de 2019

Ver no GitHub
 (0 comments) (0 reactions) (0 assignees)Python (644 forks)auto 404
? - Needs TriageCUDA / C++Cython / Pythonfeature requestgood first issue

Métricas do repositório

Stars
 (5.223 stars)
Métricas de merge de PR
 (Métricas PR pendentes)

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.

Guia do colaborador