Restricted local step function for HDPHMM- HDP-HMM with moves

[Jingyao12]

Hi
It is me again. I am trying to combine moves with HDP-HMM using following parameters for my own data

goodelbopairs_merge_kwargs = dict(
m_startLap = 2,
# Set limits to number of merges attempted each lap.
# This value specifies max number of tries for each cluster
m_maxNumPairsContainingComp = 5,
# Set "reactivation" limits
# So that each cluster is eligible again after 10 passes thru dataset
# Or when it's size changes by 400%
m_nLapToReactivate = 10,
m_minPercChangeInNumAtomsToReactivate = 400 * 0.01,
# Specify how to rank pairs (determines order in which merges are tried)
# 'obsmodel_elbo' means rank pairs by improvement to observation model ELBO
m_pair_ranking_procedure = 'obsmodel_elbo',
m_pair_ranking_direction = 'descending',
)
goodelbopairs_delete_kwargs =dict(
d_startLap=2

)

goodelbopairs_trained_model, goodelbopairs_info_dict = bnpy.run(
myData, 'HDPHMM', 'DiagGauss', 'memoVB',

moves='delete,shuffle',
**dict(
    alg_kwargs.items()
    + init_kwargs.items()
    + hdphmm_kwargs.items()
    + gauss_kwargs.items()
    + goodelbopairs_delete_kwargs.items()))

The error is as follows

Dataset Summary:
GroupXData
total size: 24 units
batch size: 24 units
num. batches: 1
Allocation Model: None
Obs. Data Model: Gaussian with diagonal covariance.
Obs. Data Prior: independent Gauss-Wishart prior on each dimension
Wishart params
nu = 22 ...
beta = [ 20 20] ...
Expectations
E[ mean[k]] =
[ 0 0] ...
E[ covar[k]] =
[[1. 0.]
[0. 1.]] ...
Initialization:
initname = randexamples
K = 50 (number of clusters)
seed = 1607680
elapsed_time: 1.7 sec
Learn Alg: memoVB | task 1/1 | alg. seed: 1607680 | data order seed: 8541952
task_output_path: /trylargefish-K=20-model=HDPHMM+DiagGauss-ECovMat=1*eye-merge_strategy=good_elbo_pairs/1
DELETE @ lap 1.00: Disabled. Cannot delete before first complete lap, because SS that represents whole dataset is required.
1.000/100 after 555 sec. | 726.0 MiB | K 50 | loss 7.068809849e-01 |

NotImplementedError Traceback (most recent call last)
in ()
29 + gauss_kwargs.items()
30 # + goodelbopairs_merge_kwargs.items()
---> 31 + goodelbopairs_delete_kwargs.items()))
32
33

/bnpy/bnpy/Run.pyc in run(dataName, allocModelName, obsModelName, algName, doSaveToDisk, doWriteStdOut, taskID, **kwargs)
92 dataName, allocModelName,
93 obsModelName, algName,
---> 94 doSaveToDisk, doWriteStdOut)
95 if (taskid == starttaskid or info_dict['loss'] < best_loss):
96 bestModel = hmodel

/bnpy/bnpy/Run.pyc in _run_task_internal(jobname, taskid, nTask, ReqArgs, KwArgs, UnkArgs, dataName, allocModelName, obsModelName, algName, doSaveToDisk, doWriteStdOut)
221
222 # Fit the model to the data!
--> 223 RunInfo = learnAlg.fit(hmodel, Data)
224 RunInfo['UnkArgs'] = UnkArgs
225 RunInfo['KwArgs'] = KwArgs

/bnpy/bnpy/learnalg/MemoVBMovesAlg.pyc in fit(self, hmodel, DataIterator, **kwargs)
118 MovePlans=MovePlans,
119 MoveRecordsByUID=MoveRecordsByUID,
--> 120 MoveLog=MoveLog)
121
122 self.saveDebugStateAtBatch(

/bnpy/bnpy/learnalg/MemoVBMovesAlg.pyc in calcLocalParamsAndSummarize_withExpansionMoves(self, DataIterator, curModel, SS, batchID, lapFrac, MovePlans, MoveRecordsByUID, MoveLog, **kwargs)
408 nUpdateSteps=DKwargs['d_nRefineSteps'],
409 d_initTargetDocTopicCount=DKwargs['d_initTargetDocTopicCount'],
--> 410 d_initWordCounts=DKwargs['d_initWordCounts'],
411 )
412 ElapsedTimeLogger.stopEvent('delete', 'localexpansion')

/bnpy/bnpy/birthmove/BRestrictedLocalStep.pyc in summarizeRestrictedLocalStep(Dslice, curModel, curLPslice, **kwargs)
30 if allocModelName not in RestrictedLocalStepFuncMap:
31 raise NotImplementedError('Restricted local step function for ' +
---> 32 allocModelName)
33 xSSslice = RestrictedLocalStepFuncMap[allocModelName](
34 Dslice=Dslice,

NotImplementedError: Restricted local step function for HDPHMM