def instantiate_model(self):
m = MultinomialModel(NDieModel(n=6))
return GaussianRandomWalkModel(m,
diagonal=False,
random_walk_idxs=[1, 2, 4],
model_transformation=(from_simplex,
to_simplex),
scale_mult='n_meas')
def instantiate_model(self):
m = MultinomialModel(NDieModel(n=6))
mult = lambda eps: eps['n_meas']**2
return GaussianRandomWalkModel(m,
diagonal=True,
random_walk_idxs=[1, 2, 4],
model_transformation=(from_simplex,
to_simplex),
scale_mult=mult)
def instantiate_model(self):
m = MultinomialModel(NDieModel(n=6))
cov = np.random.random(size=(3, 3))
cov = np.dot(cov, cov.T)
return GaussianRandomWalkModel(m,
fixed_covariance=cov,
diagonal=False,
random_walk_idxs=np.s_[:6:2],
model_transformation=(from_simplex,
to_simplex),
scale_mult='n_meas')
def test_indexing(self):
model = lambda slice: GaussianRandomWalkModel(
MultinomialModel(NDieModel(n=6)), random_walk_idxs=slice)
assert (model('all').n_modelparams == 12)
assert (model(np.s_[:6]).n_modelparams == 12)
assert (model(np.s_[:6:2]).n_modelparams == 9)
assert (model([2, 3, 4]).n_modelparams == 9)
self.assertRaises(IndexError, model, np.s_[:7])
self.assertRaises(IndexError, model, np.s_[6:])
self.assertRaises(IndexError, model, [1, 2, 8])
def instantiate_model(self):
m = BinomialModel(CoinModel())
return GaussianRandomWalkModel(m,
fixed_covariance=np.array([0.01]),
diagonal=True)