就地在float32/16中生成普通随机数 [英] Generate normal random numbers in float32/16 in-place

问题描述

在Numpy/Scipy中，如何从具有指定(浮动)dtype的正态分布生成随机数?就我而言，我需要float32和float16.

In Numpy/Scipy, how can I generate random numbers from a normal distribution with a specified (float) dtype? In my case I need float32 and float16.

由于数组很大，因此我不希望在采样后转换.

Since the array is quite large, I don't want to convert the array after the sampling.

例如:

a = np.random.normal(1e7).astype('float16')

可以完成这项工作，但是由于它需要一个临时的float64数组，因此它使用的RAM是直接float16采样的4倍.

does the job but since it need a temporary float64 array it uses 4x the RAM than a direct float16 sampling.

推荐答案

我不知道numpy或scipy中的随机数生成器会本地生成16或32位浮点数.

I don't know of a random number generator in numpy or scipy that generates 16 or 32 bit floats natively.

为避免产生较大的临时值，可以分批生成值.例如，下面的代码创建一个包含10000000个float16值样本的数组.

To avoid the large temporary, you could generate the values in batches. For example, the following creates an array of 10000000 samples of float16 values.

In [125]: n = 10000000  # Number of samples to generate

In [126]: k = 10000     # Batch size

In [127]: a = np.empty(n, dtype=np.float16)

In [128]: for i in range(0, n, k):
   .....:     a[i:i+k] = np.random.normal(loc=0, scale=1, size=k)
   .....:

这篇关于就地在float32/16中生成普通随机数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！