i encountered </s>
word/character in vocabulary created word2vec separate word.
although did tried search web character, cannot specify character @ search engines.
so, knows character is?
if @ line 82 of source code of word2vec
,
if (ch == '\n') { strcpy(word, (char *)"</s>"); return; }
</s>
character used mikolov et al. denote end of line (or more precisely \n
). don't think has special html/latex reference. nor appears on ascii chart.
Comments
Post a Comment