You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you for providing a library for Bayesian non-parametric modeling. I have been trying to use the HDPHMM with birth, and merge moves for EEG spectral data. But it seems for the GroupXData, the following error keeps occurring when trying birth moves. Can I get some help in this regard? I would really appreciate it, I have been stuck in this problem for a long time.
Dataset Summary:
GroupXData
total size: 6660 units
batch size: 6660 units
num. batches: 1
Allocation Model: None
Obs. Data Model: Gaussian with full covariance.
Obs. Data Prior: Gauss-Wishart on mean and covar of each cluster
E[ mean[k] ] =
[0. 0.] ...
E[ covar[k] ] =
[[0.1 0. ]
[0. 0.1]] ...
Initialization:
initname = randexamples
K = 1 (number of clusters)
seed = 1607680
elapsed_time: 0.5 sec
Learn Alg: memoVB | task 1/1 | alg. seed: 1607680 | data order seed: 8541952
task_output_path: HDPHMM_trymoves-K=1/1
BIRTH @ lap 1.00: Disabled. Waiting for lap >= 2 (--b_startLap).
MERGE @ lap 1.00: Disabled. Cannot plan merge on first lap. Need valid SS that represent whole dataset.
1.000/100 after 35 sec. | 2482.6 MiB | K 1 | loss 1.779313458e+00 |
MERGE @ lap 2.00: Disabled. Waiting for lap >= 5 (--m_startLap).
Hello,
Thank you for providing a library for Bayesian non-parametric modeling. I have been trying to use the HDPHMM with birth, and merge moves for EEG spectral data. But it seems for the GroupXData, the following error keeps occurring when trying birth moves. Can I get some help in this regard? I would really appreciate it, I have been stuck in this problem for a long time.
Here is the model I specified:
K1_trained_model, K1_info_dict = bnpy.run(
data, 'HDPHMM', 'Gauss', 'memoVB',
output_path='HDPHMM_trymoves-K=1/',
nLap=100, nTask=1, nBatch=1,
sF=0.1, ECovMat='eye',
K=1, initname='randexamples',
moves='birth,merge,shuffle',
m_startLap=5, b_startLap=2, b_Kfresh=4)
And here is the error I am getting:
Dataset Summary:
GroupXData
total size: 6660 units
batch size: 6660 units
num. batches: 1
Allocation Model: None
Obs. Data Model: Gaussian with full covariance.
Obs. Data Prior: Gauss-Wishart on mean and covar of each cluster
E[ mean[k] ] =
[0. 0.] ...
E[ covar[k] ] =
[[0.1 0. ]
[0. 0.1]] ...
Initialization:
initname = randexamples
K = 1 (number of clusters)
seed = 1607680
elapsed_time: 0.5 sec
Learn Alg: memoVB | task 1/1 | alg. seed: 1607680 | data order seed: 8541952
task_output_path: HDPHMM_trymoves-K=1/1
BIRTH @ lap 1.00: Disabled. Waiting for lap >= 2 (--b_startLap).
MERGE @ lap 1.00: Disabled. Cannot plan merge on first lap. Need valid SS that represent whole dataset.
1.000/100 after 35 sec. | 2482.6 MiB | K 1 | loss 1.779313458e+00 |
MERGE @ lap 2.00: Disabled. Waiting for lap >= 5 (--m_startLap).
TypeError Traceback (most recent call last)
in ()
6 K=1, initname='randexamples',
7 moves='birth,merge,shuffle',
----> 8 m_startLap=5, b_startLap=2, b_Kfresh=4)
/home/cvpia/bnpy/bnpy/Runner.pyc in run(dataName, allocModelName, obsModelName, algName, doSaveToDisk, doWriteStdOut, taskID, **kwargs)
90 dataName, allocModelName,
91 obsModelName, algName,
---> 92 doSaveToDisk, doWriteStdOut)
93 if (taskid == starttaskid or info_dict['loss'] < best_loss):
94 bestModel = hmodel
/home/cvpia/bnpy/bnpy/Runner.pyc in _run_task_internal(jobname, taskid, nTask, ReqArgs, KwArgs, UnkArgs, dataName, allocModelName, obsModelName, algName, doSaveToDisk, doWriteStdOut)
241
242 # Fit the model to the data!
--> 243 RunInfo = learnAlg.fit(hmodel, Data)
244 RunInfo['UnkArgs'] = UnkArgs
245 RunInfo['KwArgs'] = KwArgs
/home/cvpia/bnpy/bnpy/learnalg/MemoVBMovesAlg.pyc in fit(self, hmodel, DataIterator, **kwargs)
118 MovePlans=MovePlans,
119 MoveRecordsByUID=MoveRecordsByUID,
--> 120 MoveLog=MoveLog)
121
122 self.saveDebugStateAtBatch(
/home/cvpia/bnpy/bnpy/learnalg/MemoVBMovesAlg.pyc in calcLocalParamsAndSummarize_withExpansionMoves(self, DataIterator, curModel, SS, batchID, lapFrac, MovePlans, MoveRecordsByUID, MoveLog, **kwargs)
371 nBatch=self.nBatch,
372 batchPos=batchPos,
--> 373 **self.algParams['birth'])
374 ElapsedTimeLogger.stopEvent('birth', 'localexpansion')
375
/home/cvpia/bnpy/bnpy/birthmove/BCreateManyProposals.pyc in makeSummariesForManyBirthProposals(Dslice, curModel, curLPslice, curSSwhole, curSSslice, LPkwargs, newUIDs, b_targetUIDs, xSSProposalsByUID, MovePlans, MoveRecordsByUID, taskoutpath, lapFrac, batchID, batchPos, nBatch, **BArgs)
66 lapFrac=lapFrac,
67 batchID=batchID,
---> 68 **BArgs)
69 if xSSslice is not None:
70 # Proposal successful, with at least 2 non-empty clusters.
/home/cvpia/bnpy/bnpy/birthmove/BCreateOneProposal.pyc in makeSummaryForBirthProposal_HTMLWrapper(Dslice, curModel, curLPslice, **kwargs)
70 else:
71 xSSslice, DebugInfo = makeSummaryForBirthProposal(
---> 72 Dslice, curModel, curLPslice, **kwargs)
73
74 # Write output to HTML
/home/cvpia/bnpy/bnpy/birthmove/BCreateOneProposal.pyc in makeSummaryForBirthProposal(Dslice, curModel, curLPslice, curSSwhole, b_creationProposalName, targetUID, ktarget, newUIDs, LPkwargs, lapFrac, batchID, seed, b_nRefineSteps, b_debugOutputDir, b_minNumAtomsForNewComp, b_doInitCompleteLP, b_cleanupWithMerge, b_method_initCoordAscent, vocabList, **kwargs)
162 logFunc=BLogger.pprint,
163 NiterForBregmanKMeans=kwargs['b_NiterForBregmanKMeans'],
--> 164 **kwargs)
165 # EXIT EARLY: if proposal initialization fails (not enough data).
166 if xInitSStarget is None:
/home/cvpia/bnpy/bnpy/init/FromScratchBregman.pyc in initSS_BregmanDiv(Dslice, curModel, curLPslice, K, ktarget, seed, includeAllocSummary, NiterForBregmanKMeans, logFunc, **kwargs)
134 ktarget,
135 K=K,
--> 136 **kwargs)
137 if logFunc:
138 logFunc(DebugInfo['targetAssemblyMsg'])
/home/cvpia/bnpy/bnpy/init/FromScratchBregman.pyc in makeDataSubsetByThresholdResp(Data, curModel, curLP, ktarget, K, minRespForEachTargetAtom, **kwargs)
484 )
485 return DebugInfo, None, None, None, None
--> 486 targetData = Data.make_subset(example_id_list=chosenRespIDs)
487 targetX = targetData.X
488 if curLP is None:
TypeError: make_subset() got an unexpected keyword argument 'example_id_list'
The text was updated successfully, but these errors were encountered: