def setUp(self):
X = np.random.randn(100, 3)
self.Data = XData(X=X)
aPDict = dict(alpha0=1.0)
oPDict = dict(min_covar=1e-9)
self.hmodel = HModel.CreateEntireModel('EM', 'MixModel', 'ZMGauss',
aPDict, oPDict, self.Data)
def setUp(self):
X = np.random.randn(100, 3)
self.Data = XData(X=X)
aPDict = dict(alpha0=1.0)
oPDict = dict(dF=5, ECovMat='eye', sF=1.0)
self.hmodel = HModel.CreateEntireModel('VB', 'MixModel', 'ZMGauss',
aPDict, oPDict, self.Data)
def setUp(self):
PRNG = np.random.RandomState(867)
X = PRNG.randn(100,2)
self.Data = XData(X=X)
aPDict = dict(alpha0=1.0)
oPDict = dict(min_covar=1e-9)
self.hmodel = HModel.CreateEntireModel('EM','MixModel','ZMGauss', aPDict, oPDict, self.Data)
initParams = dict(initname='randexamples', seed=0, K=5)
self.hmodel.init_global_params(self.Data, **initParams)
def MakeModelWithTrueComps(self):
''' Create DPMix-Gauss model of self.Data, with K=3 *true* components.
Store result as "hmodel" in self.
'''
aDict = dict(alpha0=1.0, truncType='z')
oDict = dict(kappa=1e-7, dF=1, ECovMat='eye', sF=1e-3)
self.hmodel = HModel.CreateEntireModel('VB', 'DPMixModel', 'Gauss',
aDict, oDict, self.Data)
LP = dict(resp=self.TrueResp)
SS = self.hmodel.get_global_suff_stats(self.Data, LP)
self.hmodel.update_global_params(SS)
def MakeModelWithTrueComps(self):
''' Create hmodel with "true" components as self.Data, add to self
Also create representative suff stats SS, add to self
'''
aDict = dict(alpha0=1.0, truncType='z')
oDict = dict(kappa=1e-5, dF=1, ECovMat='eye', sF=1e-3)
self.hmodel = HModel.CreateEntireModel('VB', 'DPMixModel', 'Gauss',
aDict, oDict, self.Data)
LP = dict(resp=self.TrueResp)
SS = self.hmodel.get_global_suff_stats(self.Data, LP)
self.hmodel.update_global_params(SS)
LP = self.hmodel.calc_local_params(self.Data)
flagDict = dict(doPrecompEntropy=True, doPrecompMergeEntropy=True)
self.SS = self.hmodel.get_global_suff_stats(self.Data, LP, **flagDict)
def MakeModelWithOneComp(self):
''' Create DPMix-Gauss model of self.Data, with K=1 components.
Store result as "oneModel" in self,
with updated suff stats field "oneSS"
'''
aDict = dict(alpha0=1.0, truncType='z')
oDict = dict(kappa=1e-7, dF=1, ECovMat='eye', sF=1e-3)
self.oneModel = HModel.CreateEntireModel('VB', 'DPMixModel', 'Gauss',
aDict, oDict, self.Data)
LP = dict(resp=np.ones((self.Data.nObs, 1)))
SS = self.oneModel.get_global_suff_stats(self.Data, LP)
assert SS.x.size == self.Data.dim
assert self.oneModel.obsModel.obsPrior.D == self.Data.dim
self.oneModel.update_global_params(SS)
LP = self.oneModel.calc_local_params(self.Data)
self.oneSS = self.oneModel.get_global_suff_stats(self.Data, LP)
def setUp(self, K=7):
''' Create random data, and a K component MixModel to go with it
Call this original model "hmodel".
We copy hmodel into "modelB", and then save to file via save_model()
'''
self.K = K
PRNG = np.random.RandomState(867)
X = PRNG.randn(100,2)
self.Data = XData(X=X)
aPDict = dict(alpha0=1.0)
oPDict = dict(min_covar=1e-9)
self.hmodel = HModel.CreateEntireModel('EM','MixModel','ZMGauss',
aPDict, oPDict, self.Data)
modelB = self.hmodel.copy()
initParams = dict(initname='randexamples', seed=0, K=self.K)
modelB.init_global_params(self.Data, **initParams)
ModelWriter.save_model(modelB, '/tmp/', 'Test')
self.modelB = modelB
def MakeModelWithDuplicatedComps(self):
''' Create model with "duplicated" components"
For each true component k, we have "two" versions of it, k1 and k2.
Half of the data generated by k is assigned to k1
and the other half is assigned to k2.
These assignments are done via "dupResp" local parameter,
and so by feeding that LP into SS and update_global_params.
'''
aDict = dict(alpha0=1.0, truncType='z')
oDict = dict(kappa=1e-5, dF=1, ECovMat='eye', sF=1e-3)
self.dupModel = HModel.CreateEntireModel('VB', 'DPMixModel', 'Gauss',
aDict, oDict, self.Data)
LP = dict(resp=self.DupResp)
SS = self.hmodel.get_global_suff_stats(self.Data, LP)
self.dupModel.update_global_params(SS)
LP = self.dupModel.calc_local_params(self.Data)
flagDict = dict(doPrecompEntropy=True, doPrecompMergeEntropy=True)
self.dupSS = self.dupModel.get_global_suff_stats(
self.Data, LP, **flagDict)
def MakeHModel(self):
aPDict = dict(alpha0=1.0)
oPDict = dict(min_covar=1e-9)
self.hmodel = HModel.CreateEntireModel('EM', 'MixModel', 'ZMGauss',
aPDict, oPDict, self.Data)
def MakeHModel(self):
aPDict = dict(alpha0=1.0)
oPDict = dict(dF=5, ECovMat='eye', sF=1.0)
self.hmodel = HModel.CreateEntireModel('VB', 'MixModel', 'ZMGauss',
aPDict, oPDict, self.Data)