def __init__(self,
n_topics,
corpus,
words2vec_ny,
words2vec,
vocab_ny,
alpha=None):
self.n_dk = None
self.corpus = corpus # corpus.index2doc
self.priors = None
self.n_topics = n_topics
self.vocab_ny = vocab_ny #set([])
self.topic_params = defaultdict(dict)
self.topic_params2 = defaultdict(dict)
self.word_vec_size = words2vec.vector_size
self.alpha = alpha
self.vocab2topic = {}
self.words2vec = words2vec
self.words2vec_ny = words2vec_ny
self.solver = cholesky.Helper()
self.AVGLD = util.CalculateComplexity()
self.N_sum = 0
self.logDeterminant = []
self.z_dw = []
assert isinstance(self.n_topics, int), 'n_topic should be an integer'
def __init__(self,
num_topics,
corpus,
word_vector_filepath=None,
word_vector_model=None,
alpha=0.2,
outputfile=None,
preprocess=False):
self.doc_topic_CT = None
self.corpus = corpus
self.priors = None
self.word_vecs = {}
self.numtopics = num_topics
self.vocab = set([])
self.topic_params = defaultdict(dict)
self.wordvecFP = word_vector_filepath
self.word_vec_size = None
self.alpha = alpha
self.solver = cholesky.Helper()
self.wvmodel = word_vector_model
self.test_word_topics = defaultdict(list)
self.test_word_topic_count = defaultdict(int)
self.word_topics = {}
self.output_file_name = outputfile
self.preprocess = preprocess
def __init__(self, num_topics, corpus, word_vector_filepath=None, word_vector_model=None, run_name=str(1)):
self.doc_topic_CT = None
self.word_topics = {}
self.corpus = corpus
self.priors = None
self.word_vecs = {}
self.numtopics = num_topics
self.vocab = set([])
self.topic_params = defaultdict(dict)
self.wordvecFP = word_vector_filepath
self.word_index = {}
self.word_vec_size = None
self.alpha = 20. / self.numtopics
self.solver = cholesky.Helper()
self.wvmodel = word_vector_model
self.doc_word_counts = {}
self.run_name = run_name