numba - 如何在 Colaboratory 中使用 numba

Question

有人尝试在谷歌合作中使用 numba 吗？我只是不知道如何在这种环境中设置它。目前，我被错误困住了library nvvm not found。

Answer 1

将此代码复制到单元格中。这个对我有用。

!apt-get install nvidia-cuda-toolkit
!pip3 install numba

import os
os.environ['NUMBAPRO_LIBDEVICE'] = "/usr/lib/nvidia-cuda-toolkit/libdevice"
os.environ['NUMBAPRO_NVVM'] = "/usr/lib/x86_64-linux-gnu/libnvvm.so"

from numba import cuda
import numpy as np
import time

@cuda.jit
def hello(data):
    data[cuda.blockIdx.x, cuda.threadIdx.x] = cuda.blockIdx.x

numBlocks = 5
threadsPerBlock = 10

data = np.ones((numBlocks, threadsPerBlock), dtype=np.uint8)

hello[numBlocks, threadsPerBlock](data)

print(data)

Answer 2

我不必安装@Algis 建议的软件包，但是驱动程序的路径不同。所以我不得不做以下事情。

首先确定驱动程序的正确路径

!find / -iname 'libdevice'
!find / -iname 'libnvvm.so'

# Output:
# /usr/local/cuda-9.2/nvvm/lib64/libnvvm.so
# /usr/local/cuda-9.2/nvvm/libdevice

然后将路径设置为@Algis 描述的

import os
os.environ['NUMBAPRO_LIBDEVICE'] = "/usr/local/cuda-9.2/nvvm/libdevice"
os.environ['NUMBAPRO_NVVM'] = "/usr/local/cuda-9.2/nvvm/lib64/libnvvm.so"

Answer 3

如果你在 colab 笔记本的开头有这个块，你可以在一次简单的扫描中完成 @Stan 的工作（这也会随着 CUDA 的更新而自动更新）

import os
dev_lib_path = !find / -iname 'libdevice'
nvvm_lib_path = !find / -iname 'libnvvm.so'
assert len(dev_lib_path)>0, "Device Lib Missing"
assert len(nvvm_lib_path)>0, "NVVM Missing"
os.environ['NUMBAPRO_LIBDEVICE'] = dev_lib_path[0]
os.environ['NUMBAPRO_NVVM'] = nvvm_lib_path[0]