英文:
Random sample according to distribution
问题
如何根据概率分布从列表中返回多个唯一项目？
random.sample(population, k) 恰好是我想要的，但可惜它不接受概率分布。
random.choices(population, weights, k) 也几乎是我想要的，但它的输出可以包含来自群体的相同成员超过一次。
>>> random.choices([1, 2], [0.5, 0.5], k=2)
[1, 1]
对于这些参数，我希望输出始终是 [1, 2] 或 [2, 1]。
英文:
How can I return multiple unique items from a list, according to a probability distribution?
random.sample(population, k) is exactly what I want, but sadly it doesn't take a probability distribution.
random.choices(population, weights, k) is also almost what I want, except its output can contain the same member of the population more than once
>>> random.choices([1, 2], [0.5, 0.5], k=2)
[1, 1]
For those arguments, I would like the output to always be [1, 2] or [2, 1].
答案1
得分: 2
numpy.random.choice 可能是你想要的。你可以：
- 使用
replace=False调用它以无替换方式进行抽样 - 通过参数
p指定概率权重
英文:
numpy.random.choice is probably what you want. You can:
- call it with
replace=Falseto sample without replacement - specify probability weights through parameter
p
答案2
得分: 0
如@alexis_thual所提到的，请使用np.random.choice。以下是一个示例，展示了它如何适用于你想要的情况：
import numpy as np
np.random.choice([1, 2], size=2, replace=False, p=[0.5, 0.5])
英文:
As mentioned by @alexis_thual, use np.random.choice. Here is an example of how it would work for what you want:
import numpy as np
np.random.choice([1, 2], size=2, replace=False, p=[0.5, 0.5])
答案3
得分: -2
random.choices()函数位于Python的random模块中，允许你从一个给定的总体中生成一个带有权重的随机样本。然而，正如你提到的，它可能会在输出中产生重复的项目。
要生成一个根据概率分布的随机样本并避免重复项，你可以将random.choices()与set()结合使用，从结果中移除重复项。
以下是一个示例：
import random
population = [1, 2]
weights = [0.5, 0.5]
k = 2
sample = set(random.choices(population, weights, k))
result = list(sample)
# 如果样本中唯一项的数量少于k，重复这个过程直到达到要求
while len(result) < k:
sample = set(random.choices(population, weights, k - len(result)))
result.extend(list(sample))
print(result)
这段代码将根据给定的权重，从population列表中生成一个包含k个项目的随机样本。set()函数用于从样本中移除重复项。如果样本中唯一项的数量少于k，则代码会重复抽样的过程，直到有足够数量的唯一项为止。
希望对你有所帮助！
英文:
The random.choices() function in Python's random module allows you to generate a random sample from a population while specifying weights for each item. However, as you mentioned, it can produce duplicate items in the output.
To generate a random sample according to a probability distribution without duplicates, you can combine random.choices() with set() to remove duplicate items from the result.
Here is an example:
import random
population = [1, 2]
weights = [0.5, 0.5]
k = 2
sample = set(random.choices(population, weights, k))
result = list(sample)
# If the number of unique items in the sample is less than k, repeat the process until it matches
while len(result) < k:
sample = set(random.choices(population, weights, k - len(result)))
result.extend(list(sample))
print(result)
This code will generate a random sample of k items from the population list, based on the given weights. The set() function is used to remove any duplicate items from the sample. If the number of unique items in the sample is less than k, the code repeats the sampling process until it has enough unique items.
Hope this helps!
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论