I'm trying to just run the example in the readme and get the following error:
$ python pyhsmm/examples/hsmm.py
/usr/local/lib/python2.7/site-packages/pybasicbayes/distributions/multinomial.py:21: UserWarning: using slow sample_crp_tablecounts
warn('using slow sample_crp_tablecounts')
/usr/local/lib/python2.7/site-packages/pybasicbayes/distributions/negativebinomial.py:25: UserWarning: using slow sample_crp_tablecounts
warn('using slow sample_crp_tablecounts')
/usr/local/lib/python2.7/site-packages/pyhsmm/internals/transitions.py:17: UserWarning: using slow transition counting
warn('using slow transition counting')
This demo shows the HDP-HSMM in action. Its iterations are slower than those for
the (Sticky-)HDP-HMM, but explicit duration modeling can be a big advantage for
conditioning the prior or for discovering structure in data.
Traceback (most recent call last):
File "pyhsmm/examples/hsmm.py", line 52, in <module>
posteriormodel.resample_model()
File "/usr/local/lib/python2.7/site-packages/pyhsmm/models.py", line 442, in resample_model
self.resample_states(num_procs=num_procs)
File "/usr/local/lib/python2.7/site-packages/pyhsmm/models.py", line 467, in resample_states
s.resample()
File "/usr/local/lib/python2.7/site-packages/pyhsmm/internals/hsmm_states.py", line 325, in resample
betal, betastarl = self.messages_backwards()
File "/usr/local/lib/python2.7/site-packages/pyhsmm/internals/hsmm_states.py", line 503, in messages_backwards
from pyhsmm.internals.hsmm_messages_interface import messages_backwards_log
ImportError: No module named hsmm_messages_interface
On OSX 10.12.4 with python 2.7.11. I tried building from source and using the pip install and both give the same error.
I'm trying to just run the example in the readme and get the following error:
On OSX 10.12.4 with python 2.7.11. I tried building from source and using the pip install and both give the same error.