def __init__(self, model_name=None, storage=None, *args, **kwargs):
"""
Attributes:
storage (:obj: `skills_ml.Store`): skills_ml Store object
model (:obj: `gensim.models.doc2vec.Doc2Vec`): gensim doc2vec model.
"""
BaseEmbeddingModel.__init__(self, model_name=model_name, storage=storage)
Word2Vec.__init__(self, *args, **kwargs)
self.model_type = Word2Vec.__name__.lower()
def __init__(self, *args, **kwargs):
"""
Attributes:
storage (:obj: `skills_ml.Store`): skills_ml Store object
model (:obj: `gensim.models.doc2vec.Doc2Vec`): gensim doc2vec model.
"""
ModelStorage.__init__(self, storage=kwargs.pop('storage', None))
Word2Vec.__init__(self, *args, **kwargs)
self.model_name = ""
self._metadata = None
def __init__(self, keywords, sentences, corpus_file, corpus_worker,
corpus_chunksize, case_sensitive, size, alpha, window,
min_count, max_vocab_size, sample, seed, workers, min_alpha,
sg, hs, negative, ns_exponent, cbow_mean, iter, null_word,
trim_rule, sorted_vocab, batch_words, compute_loss,
max_final_vocab):
Sec2Vec.__init__(self, sentences, corpus_file)
KeywordCorpusFactory.__init__(self, keywords, case_sensitive,
corpus_worker)
# 20181130 Hannah Chen
self.kc = self.create(SentenceIterator(self.sentences),
corpus_chunksize)
# 20181130 LIN, Y.D.: Save all sentences for training
# self.kc = self.create(self.sentences, corpus_chunksize)
# self.kc = self.create(sentences, corpus_chunksize)
self.kv = dict(((keyword, []) for keyword in self.kc.keys()))
self.keyword_count = dict(((keyword, 0) for keyword in self.kc.keys()))
self.corpus_chunksize = corpus_chunksize
# 20181126 Hannah Chen, initialize epoch_logger
epoch_logger = EpochLogger(compute_loss)
Word2Vec.__init__(self,
corpus_file=corpus_file,
size=size,
alpha=alpha,
window=window,
min_count=min_count,
max_vocab_size=max_vocab_size,
sample=sample,
seed=seed,
workers=workers,
min_alpha=min_alpha,
sg=sg,
hs=hs,
negative=negative,
ns_exponent=ns_exponent,
cbow_mean=cbow_mean,
iter=iter,
null_word=null_word,
trim_rule=trim_rule,
sorted_vocab=sorted_vocab,
batch_words=batch_words,
compute_loss=compute_loss,
max_final_vocab=max_final_vocab,
callbacks=[epoch_logger])
def __init__(self, vocabulary_counts=None, logger_name="", **kwargs):
'''
Initialization
-----------------------------------------------------
Parameters:
min_count: 对字典做阶段，少于min_count次数的单词会被丢弃，默认值为5.
size： 隐藏层的单元数，默认值为100，推荐值为几十到几百
workers： 控制训练并行， 默认是1，worker参数只有安装了Cython后才有效，没有的话，只能使用单核
'''
logger = logging.getLogger(logger_name)
self.vocabulary_counts = None
kwargs["min_count"] = kwargs.get("min_count", 5)
kwargs["workers"] = kwargs.get("workers", cpu_count())
kwargs["size"] = kwargs.get("size", 128)
kwargs["sentences"] = kwargs.get("sentences", None)
kwargs["window"] = kwargs.get("window", 10)
kwargs["sg"] = 1
kwargs["hs"] = 1
if vocabulary_counts != None:
self.vocabulary_counts = vocabulary_counts
Word2Vec.__init__(self, **kwargs)
def __init__(self, *args, **kwargs):
ModelStorage.__init__(self, storage=kwargs.pop('storage', None))
Word2Vec.__init__(self, *args, **kwargs)
self.model_name = ""
self.model_type = "word2vec"
self._metadata = None
