Skip to content

Fix bug identified in #1817, comment 17340049 #2524

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Oct 25, 2013
Merged

Fix bug identified in #1817, comment 17340049 #2524

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b98b71b
Fix bug identified in #1817, comment 17340049
rmcgibbo Oct 15, 2013
fc3fcb8
Add inline comment
rmcgibbo Oct 24, 2013
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 9 additions & 6 deletions sklearn/hmm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -560,12 +560,15 @@ def _accumulate_sufficient_statistics(self, stats, seq, framelogprob,
stats['start'] += posteriors[0]
if 't' in params:
n_observations, n_components = framelogprob.shape
lneta = np.zeros((n_observations - 1, n_components, n_components))
lnP = logsumexp(fwdlattice[-1])
_hmmc._compute_lneta(n_observations, n_components, fwdlattice,
self._log_transmat, bwdlattice, framelogprob,
lnP, lneta)
stats["trans"] += np.exp(logsumexp(lneta, 0))
# when the sample is of length 1, it contains no transitions
# so there is no reason to update our trans. matrix estimate
if n_observations > 1:
lneta = np.zeros((n_observations - 1, n_components, n_components))
lnP = logsumexp(fwdlattice[-1])
_hmmc._compute_lneta(n_observations, n_components, fwdlattice,
self._log_transmat, bwdlattice, framelogprob,
lnP, lneta)
stats["trans"] += np.exp(logsumexp(lneta, 0))

def _do_mstep(self, stats, params):
# Based on Huang, Acero, Hon, "Spoken Language Processing",
Expand Down
10 changes: 10 additions & 0 deletions sklearn/tests/test_hmm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -312,6 +312,16 @@ def test_fit_works_on_sequences_of_different_length(self):
# ValueError: setting an array element with a sequence.
h.fit(obs)

def test_fit_with_length_one_signal(self):
obs = [self.prng.rand(10, self.n_features),
self.prng.rand(8, self.n_features),
self.prng.rand(1, self.n_features)]
h = hmm.GaussianHMM(self.n_components, self.covariance_type)
# This shouldn't raise
# ValueError: zero-size array to reduction operation maximum which has no identity
h.fit(obs)


def test_fit_with_priors(self, params='stmc', n_iter=5, verbose=False):
startprob_prior = 10 * self.startprob + 2.0
transmat_prior = 10 * self.transmat + 2.0
Expand Down