def mathematics_sparseinversecovariance_modular(data,lc): from modshogun import SparseInverseCovariance from numpy import dot sic = SparseInverseCovariance() S = dot(data,data.T) Si = sic.estimate(S,lc) return Si
def mathematics_sparseinversecovariance_modular(data, lc):
from modshogun import SparseInverseCovariance
from numpy import dot
sic = SparseInverseCovariance()
S = dot(data, data.T)
Si = sic.estimate(S, lc)
return Si
def mathematics_sparseinversecovariance_modular (data,lc):
try:
from modshogun import SparseInverseCovariance
except ImportError:
print("SparseInverseCovariance not available")
exit(0)
from numpy import dot
sic = SparseInverseCovariance()
S = dot(data,data.T)
Si = sic.estimate(S,lc)
return Si
def inverse_covariance (data,lc): from modshogun import SparseInverseCovariance from numpy import dot sic = SparseInverseCovariance() #by default cov() expects each row to represent a variable, with observations in the columns cov = np.cov(data.T) max_cov = cov.max() min_cov = cov.min() #comupute inverse conariance matrix Si = sic.estimate(cov,lc) return Si
def inverse_covariance(data, lc):
from modshogun import SparseInverseCovariance
from numpy import dot
sic = SparseInverseCovariance()
#by default cov() expects each row to represent a variable, with observations in the columns
cov = np.cov(data.T)
max_cov = cov.max()
min_cov = cov.min()
#compute inverse conariance matrix
Si = sic.estimate(cov, lc)
return Si