Python MultiChainGibbsSampler Examples

Example #1

File: multi_chain_gibbs_sampler_test.py Project: springbarley/mltk

 def setUp(self):
     model = Model(20)
     model.load('../testdata/lda_model')
     vocabulary = Vocabulary()
     vocabulary.load('../testdata/vocabulary.dat')
     self.multi_chain_gibbs_sampler = \
             MultiChainGibbsSampler(model, vocabulary, 10, 10, 5)

Example #2

File: multi_chain_gibbs_sampler_test.py Project: Ambier/python-sparselda

class MultiChainGibbsSamplerTest(unittest.TestCase):

    def setUp(self):
        model = Model(20)
        model.load('../testdata/lda_model')
        vocabulary = Vocabulary()
        vocabulary.load('../testdata/vocabulary.dat')
        self.multi_chain_gibbs_sampler = \
                MultiChainGibbsSampler(model, vocabulary, 10, 10, 5)

    def test_infer_topics(self):
        doc_tokens = []
        doc_topic_dist = self.multi_chain_gibbs_sampler.infer_topics(doc_tokens)
        self.assertEqual(0, len(doc_topic_dist))

        doc_tokens = ['apple', 'ipad']
        doc_topic_dist = self.multi_chain_gibbs_sampler.infer_topics(doc_tokens)
        print doc_topic_dist
        self.assertEqual(5, len(doc_topic_dist))
        self.assertTrue(0 in doc_topic_dist)
        self.assertEqual(0.05, doc_topic_dist[0])
        self.assertTrue(1 in doc_topic_dist)
        self.assertEqual(0.32, doc_topic_dist[1])
        self.assertTrue(3 in doc_topic_dist)
        self.assertEqual(0.14, doc_topic_dist[3])

        doc_tokens = ['apple', 'ipad', 'apple', 'null', 'nokia', 'macbook']
        doc_topic_dist = self.multi_chain_gibbs_sampler.infer_topics(doc_tokens)
        print doc_topic_dist
        self.assertEqual(6, len(doc_topic_dist))

Example #3

File: multi_chain_gibbs_sampler_test.py Project: springbarley/mltk

class MultiChainGibbsSamplerTest(unittest.TestCase):
    def setUp(self):
        model = Model(20)
        model.load('../testdata/lda_model')
        vocabulary = Vocabulary()
        vocabulary.load('../testdata/vocabulary.dat')
        self.multi_chain_gibbs_sampler = \
                MultiChainGibbsSampler(model, vocabulary, 10, 10, 5)

    def test_infer_topics(self):
        doc_tokens = []
        doc_topic_dist = self.multi_chain_gibbs_sampler.infer_topics(
            doc_tokens)
        self.assertEqual(0, len(doc_topic_dist))

        doc_tokens = ['apple', 'ipad']
        doc_topic_dist = self.multi_chain_gibbs_sampler.infer_topics(
            doc_tokens)
        print doc_topic_dist
        self.assertEqual(5, len(doc_topic_dist))
        self.assertTrue(0 in doc_topic_dist)
        self.assertEqual(0.05, doc_topic_dist[0])
        self.assertTrue(1 in doc_topic_dist)
        self.assertEqual(0.32, doc_topic_dist[1])
        self.assertTrue(3 in doc_topic_dist)
        self.assertEqual(0.14, doc_topic_dist[3])

        doc_tokens = ['apple', 'ipad', 'apple', 'null', 'nokia', 'macbook']
        doc_topic_dist = self.multi_chain_gibbs_sampler.infer_topics(
            doc_tokens)
        print doc_topic_dist
        self.assertEqual(6, len(doc_topic_dist))

Example #4

File: multi_chain_gibbs_sampler_test.py Project: Ambier/python-sparselda

 def setUp(self):
     model = Model(20)
     model.load('../testdata/lda_model')
     vocabulary = Vocabulary()
     vocabulary.load('../testdata/vocabulary.dat')
     self.multi_chain_gibbs_sampler = \
             MultiChainGibbsSampler(model, vocabulary, 10, 10, 5)