本文整理汇总了Python中Bio.MarkovModel.train_bw方法的典型用法代码示例。如果您正苦于以下问题：Python MarkovModel.train_bw方法的具体用法？Python MarkovModel.train_bw怎么用？Python MarkovModel.train_bw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Bio.MarkovModel的用法示例。
在下文中一共展示了MarkovModel.train_bw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: _create_mm
# 需要导入模块: from Bio import MarkovModel [as 别名]
# 或者: from Bio.MarkovModel import train_bw [as 别名]
def _create_mm(self, motif_num, alphabet):
try:
# Only EDeN has original_motives_list
input_motif = self.original_motives_list[motif_num - 1]
except AttributeError:
input_motif = self.motives_list[motif_num - 1]
headers, instances = [list(x) for x in zip(*input_motif)]
lengths = [len(instances[i]) for i in range(len(instances))]
median_len = int(math.ceil(np.median(lengths)))
# Hidden states for Markov Model
states = [str(i + 1) for i in range(median_len)]
print "original samples: %d" % len(instances)
print "states:", len(states)
# under sampling
if (len(instances) * len(states)) > 500:
samples = 500 / len(states)
# samples = 50 # fixed sampling
print 'sample size = %d' % samples
instances = random.sample(instances, samples)
instances = random.sample(instances, samples)
try:
mm = MarkovModel.train_bw(states=states,
alphabet=alphabet,
training_data=instances)
except RuntimeError, msg:
raise RuntimeError("Motif data is too large. " + str(msg))示例2: test_train_bw
# 需要导入模块: from Bio import MarkovModel [as 别名]
# 或者: from Bio.MarkovModel import train_bw [as 别名]
def test_train_bw(self):
random.seed(0)
states = ["0", "1", "2", "3"]
alphabet = ["A", "C", "G", "T"]
training_data = ["AACCCGGGTTTTTTT", "ACCGTTTTTTT",
"ACGGGTTTTTT", "ACCGTTTTTTTT"]
output_p_initial = array([0.2275677, 0.29655611,
0.24993822, 0.22593797])
output_p_transition = array(
[[5.16919807e-001, 3.65825814e-033, 4.83080193e-001, 9.23220689e-042],
[3.65130247e-001,
1.00000000e-300,
6.34869753e-001,
1.00000000e-300],
[8.68776164e-001,
1.02254304e-034,
1.31223836e-001,
6.21835051e-047],
[3.33333333e-301, 3.33333333e-001, 3.33333333e-301, 6.66666667e-001]])
output_p_emission = array(
[[2.02593570e-301, 2.02593570e-301, 2.02593570e-301, 1.00000000e+000],
[1.00000000e-300,
1.00000000e-300,
1.00000000e+000,
1.09629016e-259],
[3.26369779e-301,
3.26369779e-301,
3.26369779e-301,
1.00000000e+000],
[3.33333333e-001, 6.66666667e-001, 3.33333333e-301, 3.33333333e-301]])
markov_model = MarkovModel.train_bw(states, alphabet, training_data)
self.assertEqual(''.join(markov_model.states), ''.join(states))
self.assertEqual(''.join(markov_model.alphabet), ''.join(alphabet))
self.assertTrue(array_equal(
around(markov_model.p_initial, decimals=3),
around(output_p_initial, decimals=3)))
self.assertTrue(array_equal(around(
markov_model.p_transition, decimals=3),
around(output_p_transition, decimals=3)))
self.assertTrue(array_equal(around(
markov_model.p_emission, decimals=3),
around(output_p_emission, decimals=3)))