Scribd
Upload
16 views8 pages

Sampling

The document discusses various probability distributions and sampling techniques in Python including uniform, categorical, normal distributions and the central limit theorem. Random numbers are generated and histograms are plotted to illustrate the distributions. Cumulative distribution functions are also plotted to demonstrate convergence to the expected distributions as the number of samples increases.

Uploaded by

damasodra33
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
16 views8 pages

Sampling

The document discusses various probability distributions and sampling techniques in Python including uniform, categorical, normal distributions and the central limit theorem. Random numbers are generated and histograms are plotted to illustrate the distributions. Cumulative distribution functions are also plotted to demonstrate convergence to the expected distributions as the number of samples increases.

Uploaded by

damasodra33
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Sampling

In [1]: %matplotlib inline


from matplotlib import pyplot as plt
import mxnet as mx
from mxnet import nd
import numpy as np
import math

In [2]: import random


for i in range(10):
print([Link]())

0.8778660335481027
0.6273669409016372
0.1916048679612935
0.09766200492928401
0.6482876396284325
0.7252315169394271
0.38240498644322407
0.6658533396231734
0.25690347700919525
0.3610061946649761
Uniform Distribution
In [3]: for i in range(10):
print([Link](1, 100))

41
39
7
9
89
75
28
40
8
49
In [4]: counts = [Link](100)
fig, axes = [Link](2, 3, figsize=(15, 8), sharex=True)
axes = [Link](6)

for i in range(1, 1000001):


counts[[Link](0, 99)] += 1
if i in [10, 100, 1000, 10000, 100000, 1000000]:
axes[int(math.log10(i))-1].bar([Link](1, 101), counts)
[Link]()
Categorical Distribution
In [5]: # number of samples
n = 1000000
y = [Link](0, 1, n)
x = [Link](1, n+1)
# count number of occurrences and divide by the number of total draws
p0 = [Link](y < 0.35) / x
p1 = [Link](y >= 0.35) / x
In [6]: [Link](figsize=(15, 8))
[Link](x, p0)
[Link](x, p1)
[Link]()
Normal Distribution
In [7]: x = [Link](-10, 10, 0.01)
p = (1/[Link](2 * [Link])) * [Link](-0.5 * x**2)
[Link](figsize=(10, 5))
[Link](x, p)
[Link]()
Central Limit Theorem in Action
In [8]: # generate 10 random sequences of 10,000 uniformly distributed random variables
tmp = [Link](size=(10000,10))
x = 1.0 * (tmp > 0.3) + 1.0 * (tmp > 0.8)
mean = 1 * 0.5 + 2 * 0.2
variance = 1 * 0.5 + 4 * 0.2 - mean**2
print('mean {}, variance {}'.format(mean, variance))
# cumulative sum and normalization
y = [Link](1,10001).reshape(10000,1)
z = [Link](x,axis=0) / y

mean 0.9, variance 0.49


In [9]: [Link](figsize=(10,5))
for i in range(10):
[Link](y,z[:,i])

[Link](y,(variance**0.5) * [Link](y,-0.5) + mean,'r')


[Link](y,-(variance**0.5) * [Link](y,-0.5) + mean,'r')
[Link]()

You might also like