2014-10-22 14:57:59 +00:00
|
|
|
from cymem.cymem cimport Pool
|
|
|
|
|
|
2014-10-23 13:59:17 +00:00
|
|
|
from .lexeme cimport Lexeme
|
|
|
|
|
from .typedefs cimport flag_t
|
|
|
|
|
from .utf8string cimport StringStore
|
|
|
|
|
|
2014-10-22 01:55:42 +00:00
|
|
|
from thinc.typedefs cimport atom_t
|
2014-09-11 14:57:08 +00:00
|
|
|
|
|
|
|
|
|
2014-09-15 01:22:40 +00:00
|
|
|
cdef class Tokens:
|
2014-10-22 14:57:59 +00:00
|
|
|
cdef Pool mem
|
2014-10-23 13:59:17 +00:00
|
|
|
cdef StringStore _string_store
|
2014-10-22 14:57:59 +00:00
|
|
|
|
2014-10-23 13:59:17 +00:00
|
|
|
cdef Lexeme** _lex_ptr
|
2014-10-22 14:57:59 +00:00
|
|
|
cdef int* _idx_ptr
|
|
|
|
|
cdef int* _pos_ptr
|
2014-10-23 13:59:17 +00:00
|
|
|
cdef Lexeme** lex
|
2014-10-22 14:57:59 +00:00
|
|
|
cdef int* idx
|
|
|
|
|
cdef int* pos
|
|
|
|
|
|
|
|
|
|
cdef int length
|
|
|
|
|
cdef int max_length
|
2014-10-14 04:21:03 +00:00
|
|
|
|
2014-10-23 13:59:17 +00:00
|
|
|
cdef int extend(self, int i, Lexeme** lexemes, int n) except -1
|
|
|
|
|
cdef int push_back(self, int i, Lexeme* lexeme) except -1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
cdef class Token:
|
|
|
|
|
cdef StringStore _string_store
|
|
|
|
|
cdef public int i
|
|
|
|
|
cdef public int idx
|
|
|
|
|
cdef public int pos
|
|
|
|
|
|
|
|
|
|
cdef public atom_t id
|
|
|
|
|
cdef public atom_t cluster
|
|
|
|
|
cdef public atom_t length
|
|
|
|
|
cdef public atom_t lex_pos
|
|
|
|
|
cdef public atom_t lex_supersense
|
|
|
|
|
|
|
|
|
|
cdef public atom_t norm
|
|
|
|
|
cdef public atom_t shape
|
|
|
|
|
cdef public atom_t vocab10k
|
|
|
|
|
cdef public atom_t asciied
|
|
|
|
|
cdef public atom_t prefix
|
|
|
|
|
cdef public atom_t suffix
|
|
|
|
|
|
|
|
|
|
cdef public float prob
|
|
|
|
|
|
|
|
|
|
cdef public flag_t flags
|
