def prepare_stochastic_operators(M, p1, p2=0, type='L'):
I = MultiindexSet.createCompleteOrderSet(M, p1).arr
if type == 'L':
print "Legendre triple", I.shape
polysys = LegendrePolynomials()
L = evaluate_triples(polysys, I, I)
elif type == 'H':
assert False
#J = MultiindexSet.createCompleteOrderSet(M, p2).arr
#H = evaluate_Hermite_triple(I, I, J)
#L = [L[:, :, k] for k in range(L.shape[2])]
return L
from matplotlib.pyplot import figure, show, spy
from spuq.math_utils.multiindex_set import MultiindexSet
from spuq.polyquad.structure_coefficients import evaluate_triples
from spuq.polyquad.polynomials import LegendrePolynomials
I = MultiindexSet.createCompleteOrderSet(4, 3, reversed=True).arr
print I
#H = evaluate_Hermite_triple(I, I, J)
#print "shape of H:", H.shape
# create polynomial instance
lp = LegendrePolynomials()
L = evaluate_triples(lp, I, I)
print len(L), L[0].shape
# fig = figure()
# spy(np.sum(H, axis=2))
fig = figure()
#spy(np.sum(L))
spy(L[3])
show()
