def covariance_matrix(self, x, y, names=None, cov=None):
"""build a pyemu.Cov instance from Vario2d
Parameters
----------
x : iterable of floats
x locations
y : iterable of floats
y locations
names : iterable of str (optional)
names of location. If None, generic names will be used
cov : pyemu.Cov instance (optional)
an existing Cov instance to add contribution to
Returns
-------
pyemu.Cov
"""
if not isinstance(x, np.ndarray):
x = np.array(x)
if not isinstance(y, np.ndarray):
y = np.array(y)
assert x.shape[0] == y.shape[0]
if names is not None:
assert x.shape[0] == len(names)
c = np.zeros((len(names), len(names)))
np.fill_diagonal(c, self.contribution)
cov = Cov(x=c, names=names)
elif cov is not None:
assert cov.shape[0] == x.shape[0]
names = cov.row_names
c = np.zeros((len(names), 1)) + self.contribution
cont = Cov(x=c, names=names, isdiagonal=True)
cov += cont
else:
raise Exception("Vario2d.covariance_matrix() requires either" +
"names or cov arg")
for i1, (n1, x1, y1) in enumerate(zip(names, x, y)):
dx = x1 - x[i1 + 1:]
dy = y1 - y[i1 + 1:]
dxx = (dx * self.rotation_coefs[0]) +\
(dy * self.rotation_coefs[1])
dyy = ((dx * self.rotation_coefs[2]) +\
(dy * self.rotation_coefs[3])) *\
self.anisotropy
h = np.sqrt(dxx * dxx + dyy * dyy)
h[h < 0.0] = 0.0
h = self.h_function(h)
if np.any(np.isnan(h)):
raise Exception("nans in h for i1 {0}".format(i1))
cov.x[i1, i1 + 1:] += h
for i in range(len(names)):
cov.x[i + 1:, i] = cov.x[i, i + 1:]
return cov
def schur_test_nonpest():
import numpy as np
from pyemu import Matrix, Cov, Schur, Jco
#non-pest
pnames = ["p1", "p2", "p3"]
onames = ["o1", "o2", "o3", "o4"]
npar = len(pnames)
nobs = len(onames)
j_arr = np.random.random((nobs, npar))
jco = Jco(x=j_arr, row_names=onames, col_names=pnames)
parcov = Cov(x=np.eye(npar), names=pnames)
obscov = Cov(x=np.eye(nobs), names=onames)
forecasts = "o2"
s = Schur(jco=jco, parcov=parcov, obscov=obscov, forecasts=forecasts)
print(s.get_parameter_summary())
print(s.get_forecast_summary())
#this should fail
passed = False
try:
print(s.get_par_group_contribution())
passed = True
except Exception as e:
print(str(e))
if passed:
raise Exception("should have failed")
#this should fail
passed = False
try:
print(s.get_removed_obs_group_importance())
passed = True
except Exception as e:
print(str(e))
if passed:
raise Exception("should have failed")
print(s.get_par_contribution({"group1": ["p1", "p3"]}))
print(s.get_removed_obs_importance({"group1": ["o1", "o3"]}))
print(s.pandas)
forecasts = Matrix(x=np.random.random((1, npar)),
row_names=[forecasts],
col_names=pnames)
sc = Schur(jco=jco, forecasts=forecasts.T, parcov=parcov, obscov=obscov)
ffile = os.path.join("temp", "forecasts.jcb")
forecasts.to_binary(ffile)
sc = Schur(jco=jco, forecasts=ffile, parcov=parcov, obscov=obscov)
def errvar_test_nonpest():
import numpy as np
from pyemu import ErrVar, Matrix, Cov
#non-pest
pnames = ["p1","p2","p3"]
onames = ["o1","o2","o3","o4"]
npar = len(pnames)
nobs = len(onames)
j_arr = np.random.random((nobs,npar))
jco = Matrix(x=j_arr,row_names=onames,col_names=pnames)
parcov = Cov(x=np.eye(npar),names=pnames)
obscov = Cov(x=np.eye(nobs),names=onames)
forecasts = "o2"
omitted = "p3"
e = ErrVar(jco=jco,parcov=parcov,obscov=obscov,forecasts=forecasts,
omitted_parameters=omitted)
svs = [0,1,2,3,4,5]
print(e.get_errvar_dataframe(svs))
def covariance_matrix(self, x, y, names=None, cov=None):
"""build a pyemu.Cov instance from GeoStruct
Parameters
----------
x : iterable of floats
x locations
y : iterable of floats
y locations
names : iterable of str (optional)
names of location. If None, generic names will be used
cov : pyemu.Cov instance (optional)
an existing Cov instance to add contribution to
Returns
-------
pyemu.Cov
"""
if not isinstance(x, np.ndarray):
x = np.array(x)
if not isinstance(y, np.ndarray):
y = np.array(y)
assert x.shape[0] == y.shape[0]
if names is not None:
assert x.shape[0] == len(names)
c = np.zeros((len(names), len(names)))
np.fill_diagonal(c, self.nugget)
cov = Cov(x=c, names=names)
elif cov is not None:
assert cov.shape[0] == x.shape[0]
names = cov.row_names
c = np.zeros((len(names), 1))
c += self.nugget
cont = Cov(x=c, names=names, isdiagonal=True)
cov += cont
else:
raise Exception("GeoStruct.covariance_matrix() requires either " +
"names or cov arg")
for v in self.variograms:
v.covariance_matrix(x, y, cov=cov)
return cov
def covariance_matrix(self, x, y, names=None, cov=None):
if not isinstance(x, np.ndarray):
x = np.array(x)
if not isinstance(y, np.ndarray):
y = np.array(y)
assert x.shape[0] == y.shape[0]
if names is not None:
assert x.shape[0] == len(names)
c = np.zeros((len(names), len(names)))
np.fill_diagonal(c, self.contribution)
cov = Cov(x=c, names=names)
elif cov is not None:
assert cov.shape[0] == x.shape[0]
names = cov.row_names
c = np.zeros((len(names), 1)) + self.contribution
cont = Cov(x=c, names=names, isdiagonal=True)
cov += cont
else:
raise Exception("Vario2d.covariance_matrix() requires either" +
"names or cov arg")
for i1, (n1, x1, y1) in enumerate(zip(names, x, y)):
dx = x1 - x[i1 + 1:]
dy = y1 - y[i1 + 1:]
dxx = (dx * self.rotation_coefs[0]) +\
(dy * self.rotation_coefs[1])
dyy = ((dx * self.rotation_coefs[2]) +\
(dy * self.rotation_coefs[3])) *\
self.anisotropy
h = np.sqrt(dxx * dxx + dyy * dyy)
h[h < 0.0] = 0.0
cov.x[i1, i1 + 1:] += self.h_function(h)
for i in range(len(names)):
cov.x[i + 1:, i] = cov.x[i, i + 1:]
return cov
def covariance_matrix(self, x, y, names=None, cov=None):
if not isinstance(x, np.ndarray):
x = np.array(x)
if not isinstance(y, np.ndarray):
y = np.array(y)
assert x.shape[0] == y.shape[0]
if names is not None:
assert x.shape[0] == len(names)
c = np.zeros((len(names), len(names)))
np.fill_diagonal(c, self.nugget)
cov = Cov(x=c, names=names)
elif cov is not None:
assert cov.shape[0] == x.shape[0]
names = cov.row_names
c = np.zeros((len(names), 1)) + self.nugget
cont = Cov(x=c, names=names, isdiagonal=True)
cov += cont
else:
raise Exception("GeoStruct.covariance_matrix() requires either" +
"names or cov arg")
for v in self.variograms:
v.covariance_matrix(x, y, cov=cov)
return cov
def diagonal_cov_draw_test():
import os
import numpy as np
from pyemu import MonteCarlo, Cov, Pst
jco = os.path.join("pst", "pest.jcb")
pst = Pst(jco.replace(".jcb", ".pst"))
mc = MonteCarlo(jco=jco, pst=pst)
num_reals = 10
mc.draw(num_reals, obs=True)
print(mc.obsensemble)
pe1 = mc.parensemble.copy()
cov = Cov(x=mc.parcov.as_2d, names=mc.parcov.row_names)
#print(type(cov))
mc = MonteCarlo(jco=jco, pst=pst)
mc.parensemble.reseed()
mc.draw(num_reals, cov=cov)
pe2 = mc.parensemble