本文整理汇总了Python中hmmlearn.hmm.GaussianHMM._compute_log_likelihood方法的典型用法代码示例。如果您正苦于以下问题：Python GaussianHMM._compute_log_likelihood方法的具体用法？Python GaussianHMM._compute_log_likelihood怎么用？Python GaussianHMM._compute_log_likelihood使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类hmmlearn.hmm.GaussianHMM的用法示例。
在下文中一共展示了GaussianHMM._compute_log_likelihood方法的2个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: test_backward_with_hmmlearn
# 需要导入模块: from hmmlearn.hmm import GaussianHMM [as 别名]
# 或者: from hmmlearn.hmm.GaussianHMM import _compute_log_likelihood [as 别名]
def test_backward_with_hmmlearn(self):
r = np.random.randn
obs = [np.array([[-600 + r(), 100 + r()], [-300 + r(), 200 + r()], [0 + r(), 300 + r()]]) for _ in xrange(10)]
hmm = GaussianHMM(n_components=3)
hmm.fit(obs)
# Calculcate bwdlattice using hmmlearn algorithm
framelogprob = hmm._compute_log_likelihood(obs[0])
start = timeit.default_timer()
bwdlattice1 = hmm._do_backward_pass(framelogprob)
print('hmmlearn took %fs' % (timeit.default_timer() - start))
# Calculate bwdlattice using fhmm algorithm with #chains = 1. This should yield the exact same results
start = timeit.default_timer()
bwdlattice2 = np.zeros(bwdlattice1.shape)
fhmmc._backward(obs[0].shape[0], 1, hmm.n_components, [(x,) for x in xrange(hmm.n_components)],
hmm._log_startprob.reshape(1, 3), hmm._log_transmat.reshape(1, 3, 3), framelogprob, bwdlattice2)
print('fhmm took %fs' % (timeit.default_timer() - start))
self.assertTrue(np.allclose(bwdlattice1, bwdlattice2))示例2: range
# 需要导入模块: from hmmlearn.hmm import GaussianHMM [as 别名]
# 或者: from hmmlearn.hmm.GaussianHMM import _compute_log_likelihood [as 别名]
for chr in chrOrder:
lastClass[chr] = np.tile(args.ploidy,allData[chr].shape[0]);
for i in range(0,args.iterations):
warned=False;
if args.verbose>1: sys.stderr.write(" Iteration %i.\n"%(i));
viterbi = {}
noChange=0;
viterbiCat =np.zeros(allDataCat.shape[0]);
curTot=0;
curNumCNVs=0;
for chr in chrOrder:
#3. Calculate Viterbi path given data
if args.verbose>2: sys.stderr.write(" i=%i; Calculating Viterbi path for %s.\n"%(i,chr));
framelogprob = model._compute_log_likelihood(allData[chr])
#sys.stderr.write("framelogprob dim: "+str(framelogprob.shape)+"\n");
framelogprob[:,cnvsToStateIs[args.ploidy]] = np.subtract(framelogprob[:,cnvsToStateIs[args.ploidy]], args.standardPrior); #add log(prior)
if args.scalePDF>0:
framelogprob = np.subtract(framelogprob,statePDFMaxima) #### This requires some explanation. See Note 1 below.
logprob, viterbi[chr] = model._do_viterbi_pass(framelogprob);
curLen = len(viterbi[chr]);
#4. For each non-standard state, calculate the mean in that state and add a state with a mean representing that ploidy
changeStart=-1
viterbi[chr] = np.insert(viterbi[chr],[0,curLen],[normalState,normalState]); # add initial and terminal normalStates so that telomeres in CNV will be detected.
for j in range(1,len(viterbi[chr])):
if viterbi[chr][j]!=viterbi[chr][j-1]:#there was a change
if changeStart==-1:
if viterbi[chr][j]==normalState:
raise Exception("new state is normal ploidy state");
changeStart=j;