取决于数值的Numpy性能差异 [英] Numpy performance differences depending on numerical values

问题描述

在评估Numpy中的表达式时，我发现了一个奇怪的性能差异.

I found a strange performance difference while evaluating an expression in Numpy.

我执行了以下代码:

import numpy as np
myarr = np.random.uniform(-1,1,[1100,1100])

然后

%timeit np.exp( - 0.5 * (myarr / 0.001)**2 )
>> 184 ms ± 301 µs per loop (mean ± std. dev. of 7 runs, 1 loop each)

和

%timeit np.exp( - 0.5 * (myarr / 0.1)**2 )
>> 12.3 ms ± 34.3 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)

在第二种情况下，计算速度提高了将近15倍！注意，唯一的区别是系数为0.1或0.001.

That's an almost 15x faster computation in the second case! Note that the only difference is the factor being 0.1 or 0.001.

这种行为的原因是什么?我可以更改某些内容以使第一个计算与第二个一样快吗?

What's the reason for this behaviour? Can I change something to make the first calculation as fast as the second?

推荐答案

这可能会产生非规范化的数字，从而减慢计算速度.

This may produce denormalised numbers which slow down computations.

您可能想使用 daz 库来禁用非规范化的数字:

You may like to disable denormalized numbers using daz library:

import daz
daz.set_daz()

更多信息:

为避免由于异常和下溢次数导致的序列化和性能问题，请使用SSE和SSE2指令在硬件中设置齐平为零"和归零为零"模式，以便为浮点应用程序实现最高性能. /p>

To avoid serialization and performance issues due to denormals and underflow numbers, use the SSE and SSE2 instructions to set Flush-to-Zero and Denormals-Are-Zero modes within the hardware to enable highest performance for floating-point applications.

请注意，在64位模式下，浮点计算使用SSE指令，而不是x87.

Note that in 64-bit mode floating point computations use SSE instructions, not x87.

这篇关于取决于数值的Numpy性能差异的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！