def test():
# test single distribution
d0 = ss.uniform(loc=-1, scale=2) # uniform distribution,low=-1, width=2
print(lhd(dist=d0, size=5))
# test single distribution for multiple variables
d1 = ss.norm(loc=0, scale=1) # normal distribution, mean=0, stdev=1
print(lhd(dist=d1, size=7, dims=5))
# test multiple distributions
d2 = ss.beta(2, 5) # beta distribution, alpha=2, beta=5
d3 = ss.expon(scale=1 / 1.5) # exponential distribution, lambda=1.5
print(lhd(dist=(d1, d2, d3), size=6))
rand_lhs = lhd(dist=(d0, d1, d2, d3), size=100)
spac_lhs = lhd(
dist=(d0, d1, d2, d3),
size=100,
form="spacefilling",
iterations=100,
showcorrelations=True,
)
pytest.importorskip("matplotlib")
pytest.importorskip("scatterplot_matrix")
try:
from scatterplot_matrix import scatterplot_matrix as spm
import matplotlib.pyplot as plt
except ImportError:
print(rand_lhs)
print(spac_lhs)
else:
names = ["U(-1,1)", "N(0,1)", "Beta(2,5)", "Exp(1.5)"]
spm(rand_lhs.T, names=names)
plt.suptitle("Completely Random LHS Design")
plt.show()
spm(spac_lhs.T, names=names)
plt.suptitle("Space-Filling LHS Design")
plt.show()
# test single distribution for multiple variables
d1 = ss.norm(loc=0, scale=1) # normal distribution, mean=0, stdev=1
print(lhd(dist=d1, size=7, dims=5))
# test multiple distributions
d2 = ss.beta(2, 5) # beta distribution, alpha=2, beta=5
d3 = ss.expon(scale=1 / 1.5) # exponential distribution, lambda=1.5
print(lhd(dist=(d1, d2, d3), size=6))
rand_lhs = lhd(dist=(d0, d1, d2, d3), size=100)
spac_lhs = lhd(dist=(d0, d1, d2, d3),
size=100,
form='spacefilling',
iterations=100,
showcorrelations=True)
try:
from scatterplot_matrix import scatterplot_matrix as spm
import matplotlib.pyplot as plt
except ImportError:
print(rand_lhs)
print(spac_lhs)
else:
names = ['U(-1,1)', 'N(0,1)', 'Beta(2,5)', 'Exp(1.5)']
spm(rand_lhs.T, names=names)
plt.suptitle('Completely Random LHS Design')
plt.show()
spm(spac_lhs.T, names=names)
plt.suptitle('Space-Filling LHS Design')
plt.show()
d0 = ss.uniform(loc=-1,scale=2) # uniform distribution,low=-1, width=2
print lhd(dist=d0,size=5)
# test single distribution for multiple variables
d1 = ss.norm(loc=0,scale=1) # normal distribution, mean=0, stdev=1
print lhd(dist=d1,size=7,dims=5)
# test multiple distributions
d2 = ss.beta(2,5) # beta distribution, alpha=2, beta=5
d3 = ss.expon(scale=1/1.5) # exponential distribution, lambda=1.5
print lhd(dist=(d1,d2,d3),size=6)
rand_lhs = lhd(dist=(d0,d1,d2,d3),size=100)
spac_lhs = lhd(dist=(d0,d1,d2,d3),size=100,form='spacefilling',
iterations=100,showcorrelations=True)
try:
from scatterplot_matrix import scatterplot_matrix as spm
import matplotlib.pyplot as plt
except ImportError:
print rand_lhs
print spac_lhs
else:
names = ['U(-1,1)','N(0,1)','Beta(2,5)','Exp(1.5)']
spm(rand_lhs.T,names=names)
plt.suptitle('Completely Random LHS Design')
plt.show()
spm(spac_lhs.T,names=names)
plt.suptitle('Space-Filling LHS Design')
plt.show()