Question

我正在使用高斯LDA，并且在运行 g.fit（2）时遇到不知道如何处理uri 。该问题应与已定义的函数 fit（self，erations = 1，init = True）有关。以下是错误。

TypeError                                 Traceback (most recent call last)
<ipython-input-6-2f3f6e257f87> in <module>
     13     model_t = KeyedVectors.load_word2vec_format(tmp_file)
     14     g = Gauss_LDA(2, title_document, model_t)
---> 15     g.fit(2) #bug appears here

<ipython-input-3-89d3f37594a9> in fit(self, iterations, init)
     64     def fit(self, iterations=1, init=True): 
     65         if init == True:
---> 66             self.init()
     67             init = False

<ipython-input-3-89d3f37594a9> in init(self)
     76         self.process_corpus(self.corpus)
---> 77         self.process_wordvectors(self.wordvecFP)
     78         #setting wishhart priors
     79         self.priors = Wishart(self.word_vecs)

<ipython-input-3-89d3f37594a9> in process_wordvectors(self, filepath)
---> 37         vectors = gensim.models.KeyedVectors.load_word2vec_format(fname=filepath, binary=False)

/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/gensim/models/keyedvectors.py in load_word2vec_format(cls, fname, fvocab, binary, encoding, unicode_errors, limit, datatype)
   1474         return _load_word2vec_format(
   1475             cls, fname, fvocab=fvocab, binary=binary, encoding=encoding, unicode_errors=unicode_errors,
-> 1476             limit=limit, datatype=datatype)
   1477 
   1478     def get_keras_embedding(self, train_embeddings=False):

/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/gensim/models/utils_any2vec.py in _load_word2vec_format(cls, fname, fvocab, binary, encoding, unicode_errors, limit, datatype)
    341     logger.info("loading projection weights from %s", fname)
--> 342     with utils.smart_open(fname) as fin:
    343         header = utils.to_unicode(fin.readline(), encoding=encoding)
    344         vocab_size, vector_size = (int(x) for x in header.split())  # throws for invalid file format

/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/smart_open/smart_open_lib.py in smart_open(uri, mode, **kw)
    229     except KeyError:
    230         binary_mode = mode
--> 231     binary, filename = _open_binary_stream(uri, binary_mode, **kw)
    232     if ignore_extension:
    233         decompressed = binary

/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/smart_open/smart_open_lib.py in _open_binary_stream(uri, mode, **kw)
    378         return uri, filename
    379     else:
--> 380         raise TypeError('don\'t know how to handle uri %s' % repr(uri))

TypeError: don't know how to handle uri <gensim.models.keyedvectors.Word2VecKeyedVectors object at 0x10a80d128>

我引用此站点来修改我的代码：https://github.com/mansweet/Gaussian-LDA-word2vec 而且我发现原始代码在输入数据格式方面存在问题，因此，我通过以下方式将手套格式转换为word2vec格式：https://radimrehurek.com/gensim/scripts/glove2word2vec.html?fbclid=IwAR3ZAkpX795pyRR-XMnDYb7EW5qTNZpcZqKxTQcDkfdOqMDQO2uG-jumaWg

现在，g.fit（2）出现了问题。我查看了网站，但没有看到任何类似的问题。最接近的是：https://github.com/mansweet/Gaussian-LDA-word2vec/issues/2。

不知道如何处理uri 出现在最后一行代码中：

if __name__ == "__main__":

wordvec_fileapth = "/Users/julia/Desktop/homework/android.en.vector"
tmp_file = get_tmpfile("test_word2vec.txt")

_ = glove2word2vec(wordvec_fileapth, tmp_file)

model_t = KeyedVectors.load_word2vec_format(tmp_file) 

g = Gauss_LDA(2, title_document, model_t)

g.fit(2) ###bug appears here

错误回溯应该与以下代码相关（3个问题###）：

 def fit(self, iterations=1, init=True): 
    if init == True:
        self.init() ### traceback problem one
        init = False

    for i in xrange(iterations):
        self.sample()



def init(self):
    self.process_corpus(self.corpus)
    self.process_wordvectors(self.wordvecFP) ###traceback problem two
    #setting wishhart priors
    self.priors = Wishart(self.word_vecs)
    self.doc_topic_CT = np.zeros((len(self.corpus.keys()), self.numtopics))

    self.word_topics = {word: random.choice(range(self.numtopics)) for word in self.vocab}
    # get Doc-Topic Counts
    for docID, doc in self.corpus.iteritems():
        for word in doc:
            topicID = self.word_topics[word]
            self.doc_topic_CT[docID, topicID] += 1

    # Init parameters for topic distributions
    for k in range(self.numtopics):
        self.recalculate_topic_params(k)

def process_wordvectors(self, filepath):
    print("Processing word-vectors, this takes a moment") 

    vectors = gensim.models.KeyedVectors.load_word2vec_format(fname=filepath, binary=False) ###traceback problem three

    useable_vocab = 0
    unusable_vocab = 0
    self.word_vec_size = vectors.vector_size

    for word in self.vocab:
        try:
            vectors[word]
            useable_vocab += 1
        except KeyError: unusable_vocab += 1

    # self.word_vecs = np.zeros((useable_vocab, vectors.vector_size))
    index = 0
    for word in self.vocab:
        try:
            self.word_vecs[word] = vectors[word]
            index += 1
        except KeyError: continue
    print ("Word-vectors for the corpus are created")

我希望可以通过提供任何想法或指导来解决g.fit（）问题的原因。谢谢！

修改后的LDA出现在“ fit（）”中，“不知道如何处理uri”

0 个答案: