def testWithMixgaussInit(self):
LL, prior1, transmat1, mu1, Sigma1, mixmat1 = mhmm_em(data=self.obs,
prior=self.prior0,
transmat=self.transmat0,
mu=self.mu0,
Sigma=self.Sigma0,
mixmat=self.mixmat0)
def testStateHiddenParams(self):
LL, prior1, transmat1, mu1, Sigma1, mixmat1 = mhmm_em(data=self.obs, prior=self.prior0, transmat=self.transmat0, mu=self.mu, Sigma=self.Sigma,
mixmat=self.mixmat)
assert np.all(np.diff(LL)>=-1e-3) # Likelihood is increasing
assert np.all(np.abs(self.prior-prior1)<1e-3)
assert np.all(np.abs(self.transmat-transmat1)<1e-3)
def testWithMixgaussInit(self):
LL, prior1, transmat1, mu1, Sigma1, mixmat1 = mhmm_em(
data=self.obs,
prior=self.prior0,
transmat=self.transmat0,
mu=self.mu0,
Sigma=self.Sigma0,
mixmat=self.mixmat0)
def testStateEmissionParams(self):
LL, prior1, transmat1, mu1, Sigma1, mixmat1 = mhmm_em(data=self.obs, prior=self.prior, transmat=self.transmat, mu=self.mu0, Sigma=self.Sigma0,
mixmat=self.mixmat0)
assert np.all(np.diff(LL)>=-1e-3) # Likelihood is increasing
assert np.all(np.abs(mu1-self.mu)<1e-1)
assert np.all(np.abs(Sigma1-self.Sigma)<1e-1)
assert np.all(np.abs(mixmat1-self.mixmat)<1e-1)
def testStateHiddenParams(self):
LL, prior1, transmat1, mu1, Sigma1, mixmat1 = mhmm_em(
data=self.obs,
prior=self.prior0,
transmat=self.transmat0,
mu=self.mu,
Sigma=self.Sigma,
mixmat=self.mixmat)
assert np.all(np.diff(LL) >= -1e-3) # Likelihood is increasing
assert np.all(np.abs(self.prior - prior1) < 1e-3)
assert np.all(np.abs(self.transmat - transmat1) < 1e-3)
def testStateEmissionParams(self):
LL, prior1, transmat1, mu1, Sigma1, mixmat1 = mhmm_em(
data=self.obs,
prior=self.prior,
transmat=self.transmat,
mu=self.mu0,
Sigma=self.Sigma0,
mixmat=self.mixmat0)
assert np.all(np.diff(LL) >= -1e-3) # Likelihood is increasing
assert np.all(np.abs(mu1 - self.mu) < 1e-1)
assert np.all(np.abs(Sigma1 - self.Sigma) < 1e-1)
assert np.all(np.abs(mixmat1 - self.mixmat) < 1e-1)
def fit(self, obs):
obs = self._convertObs(obs)
O = obs[0].shape[0]
M = self.n_mix
Q = self.n_components
if 's' in self.init_params:
self.startprob_, _ = normalise(self.startprob_)
if 't' in self.init_params:
self.transmat_, _ = mk_stochastic(self.transmat_)
if 'm' in self.init_params or 'c' in self.init_params:
mu0, Sigma0, weights0 = mixgauss_init(
Q * M, obs, cov_type=self._covariance_type)
if 'm' in self.init_params:
self.means_ = np.transpose(np.reshape(mu0, (O, M, Q)),
(0, 2, 1))
if 'c' in self.init_params:
self.covars_ = np.transpose(np.reshape(Sigma0, (O, O, M, Q)),
(0, 1, 3, 2))
mixmat0, _ = mk_stochastic(np.random.rand(Q, M))
self.LL, prior1, transmat1, mu1, Sigma1, mixmat1 = mhmm_em(
data=obs,
prior=self.startprob_,
transmat=self.transmat_,
mu=self.means_,
Sigma=self.covars_,
mixmat=mixmat0,
max_iter=self.n_iter,
thresh=self.thresh,
cov_type=self._covariance_type,
adj_trans='t' in self.params,
adj_mix='w' in self.params,
adj_mu='m' in self.params,
adj_Sigma='c' in self.params)
self.startprob_ = prior1
self.transmat_ = transmat1
self.means_ = mu1
self.covars_ = Sigma1
self.weights_ = mixmat1
def fit(self, obs):
obs = self._convertObs(obs)
O = obs[0].shape[0]
M = self.n_mix
Q = self.n_components
if "s" in self.init_params:
self.startprob_, _ = normalise(self.startprob_)
if "t" in self.init_params:
self.transmat_, _ = mk_stochastic(self.transmat_)
if "m" in self.init_params or "c" in self.init_params:
mu0, Sigma0, weights0 = mixgauss_init(Q * M, obs, cov_type=self._covariance_type)
if "m" in self.init_params:
self.means_ = np.transpose(np.reshape(mu0, (O, M, Q)), (0, 2, 1))
if "c" in self.init_params:
self.covars_ = np.transpose(np.reshape(Sigma0, (O, O, M, Q)), (0, 1, 3, 2))
mixmat0, _ = mk_stochastic(np.random.rand(Q, M))
self.LL, prior1, transmat1, mu1, Sigma1, mixmat1 = mhmm_em(
data=obs,
prior=self.startprob_,
transmat=self.transmat_,
mu=self.means_,
Sigma=self.covars_,
mixmat=mixmat0,
max_iter=self.n_iter,
thresh=self.thresh,
cov_type=self._covariance_type,
adj_trans="t" in self.params,
adj_mix="w" in self.params,
adj_mu="m" in self.params,
adj_Sigma="c" in self.params,
)
self.startprob_ = prior1
self.transmat_ = transmat1
self.means_ = mu1
self.covars_ = Sigma1
self.weights_ = mixmat1
