[열유체공학실험] HW4-numba,jit,parallel,cuda를 사용시의 연산속도 비교
- 최초 등록일
- 2021.11.17
- 최종 저작일
- 2020.11
- 5페이지/
MS 워드
- 가격 1,000원
![할인쿠폰받기](/images/v4/document/ico_det_coupon.gif)
소개글
열유체공학실험 "HW4-numba,jit,parallel,cuda를 사용시의 연산속도 비교"의 보고서를 워드파일로 작성하였습니다.
목차
없음
본문내용
<CODE>
import math
import numpy as np
import matplotlib.pyplot as plt
import numba
from numba import vectorize, float32, cuda
!find / -iname 'libdevice'
!find / -iname 'libnvvm.so'
/usr/local/cuda-10.0/nvvm/libdevice
/usr/local/cuda-10.1/nvvm/libdevice
/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so
/usr/local/cuda-10.1/nvvm/lib64/libnvvm.so
import os
##################################### TO DO ###########################################
os.environ['NUMBAPRO_LIBDEVICE'] = "/usr/local/cuda-10.0/nvvm/libdevice"
os.environ['NUMBAPRO_NVVM'] = "/usr/local/cuda-10.0/nvvm/lib64/libnvvm.so"
####################################### end ###########################################
def calculate_numpy(x):
return x*2 + 3
##################################### TO DO ###########################################
@numba.jit("f4[:](f4[:])")
####################################### end ###########################################
def calculate_jit(x):
return x*2 + 3
참고 자료
없음