def test_validate_similarity_hellinger():
validator = uq.comparison.validate.ValidateSimilarityHellinger()
assert(validator.element_name() == 'validate_similarity_hellinger')
assert(validator.element_version() == '0.1')
d1 = cp.Exponential(1)
d2 = cp.Exponential(2)
xmin = min(d1.lower[0], d2.lower[0])
xmax = max(d1.upper[0], d2.upper[0])
x = np.linspace(xmin, xmax, 100)
p1 = d1.pdf(x)
p2 = d2.pdf(x)
distance = validator.compare(p1, p2)
err = abs(distance - np.sqrt(1 - 2 * np.sqrt(2) / 3))
assert err < 1.e-2
def test_dependent_density():
"""Assert that manually create dependency structure holds."""
distribution1 = chaospy.Exponential(1)
distribution2 = chaospy.Uniform(lower=0, upper=distribution1)
distribution = chaospy.J(distribution1, distribution2)
assert distribution.pdf([0.5, 0.6]) == 0
assert distribution.pdf([0.5, 0.4]) > 0
def make_descriptive():
numpy.random.seed(1234)
dist1 = chaospy.Normal(0, 1)
samples1 = dist1.sample(40)
dist2 = chaospy.Exponential()
samples2 = dist2.sample(20)
x = y = numpy.linspace(0, 2*numpy.pi, 200)
x, y = numpy.cos(x), numpy.sin(y)
pyplot.pie([0.5], colors=[COLOR1], radius=1, normalize=False,
center=(-0.3, 0.3), startangle=45,
wedgeprops={"width": 0.5, "alpha": 0.5, "lw": 4})
pyplot.plot(x-0.3, y+0.3, COLOR1, lw=4)
pyplot.plot(x/2-0.3, y/2+0.3, COLOR1, lw=4)
pyplot.bar([0, 0.6], [0.5, 1], bottom=[-0.6, -0.6],
width=0.5, yerr=[0.2, 0.3], color=COLOR2)
save("descriptive")
def test_dependent_pdf():
distribution1 = chaospy.Exponential(1)
distribution2 = chaospy.Uniform(lower=0, upper=distribution1)
distribution = chaospy.J(distribution1, distribution2)
assert distribution.pdf([0.5, 0.6]) == 0
assert distribution.pdf([0.5, 0.4]) > 0
import pytest
import chaospy
from chaospy.recurrence import RECURRENCE_ALGORITHMS
ANALYTICAL_DISTRIBUTIONS = {
"beta": chaospy.Beta(4, 2),
"expon": chaospy.Exponential(1),
"gamma": chaospy.Gamma(2, 2),
"lognorm": chaospy.LogNormal(-10, 0.1),
"normal": chaospy.Normal(2, 3),
"student": chaospy.StudentT(df=25, mu=0.5),
"uniform": chaospy.Uniform(-1, 2),
}
@pytest.fixture(params=RECURRENCE_ALGORITHMS)
def recurrence_algorithm(request):
"""Parameterization of name of recurrence algorithms."""
yield request.param
@pytest.fixture(params=ANALYTICAL_DISTRIBUTIONS.keys())
def analytical_distribution(request):
"""Parameterization of distribution with analytical TTR methods."""
return ANALYTICAL_DISTRIBUTIONS[request.param]
import pylab as pp
import numpy as np
import chaospy as cp
import os
Directory = 'Projects/WDSA_Models/'
FileName = '5Pipes.inp'
maxTime = 30
dt =0.01
Transient_Times = np.arange(0,maxTime,dt)
distribution = cp.J(cp.Normal(1,0.4))#,cp.Normal(1,0.4),cp.Normal(1,0.4),cp.Normal(1,0.4),cp.Normal(1,0.4))
demand_distribution = cp.Exponential(1.)
Iterations = 1000
#samples = np.load(Directory + '5_pipes_varying_demand_samples.npy')
#output = np.load(Directory + '5_pipes_varying_demand.npy')
samples = np.load(Directory + '5_pipes_varying_friction_all_same_samples.npy')
output = np.load(Directory + '5_pipes_varying_friction_all_same.npy')
####
# Plotting input distributions
#pp.figure()
#pp.hist(samples[0,:],density=True,color='k',bins=40,alpha=0.50,label='Samples')
#d = np.linspace(min(samples[0,:]),max(samples[0,:]),1000)
#pp.plot(d,cp.Normal(1,0.4).pdf(d),'k',label='Roughness PDF')
#pp.xlabel('Roughness (mm)')
#pp.ylabel('Frequency')
axes[1, 2].set_xlabel("N = " + str(c))
axes[1, 2].xaxis.set_label_position('top')
# 2nd row 4th plot
axes[1, 3].hist(samples2c, bins = f)
axes[1, 3].set_xlabel("N = " + str(c))
axes[1, 3].xaxis.set_label_position('top')
# 2nd row 5th plot
axes[1, 4].hist(samples2e, bins = f)
axes[1, 4].set_xlabel("N = " + str(e))
axes[1, 4].xaxis.set_label_position('top')
# Creating Exponential Distribution and samplesizes
db3 = cp.Exponential(1, 0)
samples3a = db3.sample(a, rule = 'S')
samples3b = db3.sample(b, rule = 'S')
samples3c = db3.sample(c, rule = 'S')
samples3d = db3.sample(d, rule = 'S')
samples3e = db3.sample(e, rule = 'S')
# 3rd row 1st plot
axes[2, 0].hist(samples3a, bins = f)
axes[2, 0].set_xlabel("N = " + str(a))
axes[2, 0].xaxis.set_label_position('top')
axes[2, 0].set_ylabel("Uniform")
# 3rd row 2nd plot
axes[2, 1].hist(samples3b, bins = f)
axes[2, 1].set_xlabel("N = " + str(b))
@author: anecaise
"""
import numpy as np
from matplotlib import pyplot as plt
import chaospy
# Create Fig
fig = plt.figure(figsize=[5,3], num=100)
"""exponential"""
dist = chaospy.Exponential()
x = dist.sample(10, 'S')
axes = fig.add_subplot(3,5, 1)
# axes.set_ylabel('Y Values')
axes.hist(x,bins=7,edgecolor='black')
axes.set_ylabel("Exponential")
axes.set_title('N=10')
axes.set_xticks([])
axes.set_yticks([])
x = dist.sample(50, 'S')
axes = fig.add_subplot(3,5, 2)
# axes.set_ylabel('Y Values')
axes.hist(x,bins=7,edgecolor='black')
plt.title("N=2000")
plt.yticks([])
plt.xticks([])
plt.subplot(3, 5, 5)
uniform = cp.J(cp.Uniform(0, 1))
x5k = uniform.sample(5000, rule="S")
plt.hist(x5k, density=True, bins=100)
plt.title("N=5000")
plt.yticks([])
plt.xticks([])
#Exponential Distribution
plt.subplot(3, 5, 6)
uniform = cp.J(cp.Exponential())
x100 = uniform.sample(100, rule="S")
plt.hist(x100, density=True, bins=100)
plt.ylabel("Q_Exponential")
plt.yticks([])
plt.xticks([])
plt.subplot(3, 5, 7)
uniform = cp.J(cp.Exponential())
x500 = uniform.sample(500, rule="S")
plt.hist(x500, density=True, bins=100)
plt.yticks([])
plt.xticks([])
plt.subplot(3, 5, 8)
uniform = cp.J(cp.Exponential())