def identify_outliers(scale_errors,needs_sort,bin,x):
# Useful constants
roottwo = 1.414213562373095
madfac = 1.482602218505602
n = len(x)
# Get the residuals
y = median_model(needs_sort,bin,x)
res = []
for i in range(n):
res.append(y[i][1] - x[i][1])
# Scale the errors
if scale_errors == 1:
maderror = madfac*median_deviation(res)
errors = [row[2] for row in x]
mederror = median_(errors)
sigmafac = maderror/mederror
print 'Scaling errors by ',sigmafac
else:
sigmafac = 1.0
# Number of sigmas to clip
sigmas = coolmath.inverf(1.0-(1.0/n))
# Identify outliers
outlier=[0 for i in range(n)]
for i in range(n):
if abs(res[i]) >= sigmas*sigmafac*x[i][2]:
outlier[i] = 1
else:
outlier[i] = 0
print 'Number of outliers = ',sum(outlier),' [',100.0*sum(outlier)/n,'%]'
return outlier
def identify_outliers(scale_errors, needs_sort, bin, x):
# Useful constants
roottwo = 1.414213562373095
madfac = 1.482602218505602
n = len(x)
# Get the residuals
y = median_model(needs_sort, bin, x)
res = []
for i in range(n):
res.append(y[i][1] - x[i][1])
# Scale the errors
if scale_errors == 1:
maderror = madfac * median_deviation(res)
errors = [row[2] for row in x]
mederror = median_(errors)
sigmafac = maderror / mederror
print 'Scaling errors by ', sigmafac
else:
sigmafac = 1.0
# Number of sigmas to clip
sigmas = coolmath.inverf(1.0 - (1.0 / n))
# Identify outliers
outlier = [0 for i in range(n)]
for i in range(n):
if abs(res[i]) >= sigmas * sigmafac * x[i][2]:
outlier[i] = 1
else:
outlier[i] = 0
print 'Number of outliers = ', sum(
outlier), ' [', 100.0 * sum(outlier) / n, '%]'
return outlier
def truncatedgaussian_sample(xmin, xmax, mu, sigma, x):
roottwo = 1.414213562373095
temp = math.erfc((-xmin + mu) / (roottwo * sigma)) + x * (
math.erfc((-xmax + mu) / (roottwo * sigma)) - math.erfc((-xmin + mu) / (roottwo * sigma))
)
y = mu + roottwo * sigma * coolmath.inverf(-1.0 + temp)
return y
import math import coolmath print 'x = ?' x = float(raw_input()) y = coolmath.inverf(x) print 'inverf(x) = ',y print 'erf[inverf(x)] = ',math.erf(y)
import math import coolmath print 'x = ?' x = float(raw_input()) y = coolmath.inverf(x) print 'inverf(x) = ', y print 'erf[inverf(x)] = ', math.erf(y)
def cappedgaussian_sample(xmin, mu, sigma, x):
roottwo = 1.414213562373095
temp = math.erfc((-xmin + mu) / (roottwo * sigma)) + x * (2.0 - math.erfc((-xmin + mu) / (roottwo * sigma)))
y = mu + roottwo * sigma * coolmath.inverf(-1.0 + temp)
return y
def gaussian_sample(mu, sigma, x):
roottwo = 1.414213562373095
y = mu + roottwo * sigma * coolmath.inverf(-1.0 + 2.0 * x)
return y
