Merge branch 'release-1.0.1'

Author: tmylk

gensim/models/keyedvectors.py

@@ -455,14 +455,21 @@ def most_similar_cosmul(self, positive=[], negative=[], topn=10):
             # allow calls like most_similar_cosmul('dog'), as a shorthand for most_similar_cosmul(['dog'])
             positive = [positive]
 
+        all_words = set([self.vocab[word].index for word in positive+negative
+            if not isinstance(word, ndarray) and word in self.vocab])
+
+        positive = [
+            self.word_vec(word, use_norm=True) if isinstance(word, string_types) else word
+            for word in positive
+        ]
+        negative = [
+            self.word_vec(word, use_norm=True) if isinstance(word, string_types) else word
+            for word in negative
+        ]
 
-        positive = [self.word_vec(word, use_norm=True) for word in positive]
-        negative = [self.word_vec(word, use_norm=True) for word in negative]
         if not positive:
             raise ValueError("cannot compute similarity with no input")
 
-        all_words = set([self.vocab[word].index for word in positive+negative if word in self.vocab])
-
         # equation (4) of Levy & Goldberg "Linguistic Regularities...",
         # with distances shifted to [0,1] per footnote (7)
         pos_dists = [((1 + dot(self.syn0norm, term)) / 2) for term in positive]

gensim/models/word2vec.py

@@ -1291,14 +1291,15 @@ def load(cls, *args, **kwargs):
         return model
 
     def _load_specials(self, *args, **kwargs):
+        super(Word2Vec, self)._load_specials(*args, **kwargs)
         # loading from a pre-KeyedVectors word2vec model
         if not hasattr(self, 'wv'):
             wv = KeyedVectors()
             wv.syn0 = self.__dict__.get('syn0', [])
+            wv.syn0norm = self.__dict__.get('syn0norm', None)
             wv.vocab = self.__dict__.get('vocab', {})
             wv.index2word = self.__dict__.get('index2word', [])
             self.wv = wv
-        super(Word2Vec, self)._load_specials(*args, **kwargs)
 
     @classmethod
     def load_word2vec_format(cls, fname, fvocab=None, binary=False, encoding='utf8', unicode_errors='strict',