explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-12-24 17:06:29 +03:00
Code Issues Packages Projects Releases Wiki Activity

* Add a type declaration for doc.from_array

Browse Source
This commit is contained in:
Matthew Honnibal 2015-07-27 22:57:22 +02:00
parent 25a8774f42
commit aa7a964a4f
1 changed files with 4 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
spacy/tokens/doc.pyx
View File

@ -1,5 +1,6 @@
cimport cython cimport cython
from libc.string cimport memcpy, memset from libc.string cimport memcpy, memset
from libc.stdint cimport uint32_t
import numpy import numpy
import struct import struct
@ -274,6 +275,7 @@ cdef class Doc:
cdef attr_id_t attr_id cdef attr_id_t attr_id
cdef TokenC* tokens = self.data cdef TokenC* tokens = self.data
cdef int length = len(array) cdef int length = len(array)
cdef attr_t[:] values
for col, attr_id in enumerate(attrs): for col, attr_id in enumerate(attrs):
values = array[:, col] values = array[:, col]
if attr_id == HEAD: if attr_id == HEAD:
@ -296,7 +298,8 @@ cdef class Doc:
def to_bytes(self): def to_bytes(self):
byte_string = self.vocab.serializer.pack(self) byte_string = self.vocab.serializer.pack(self)
return struct.pack('I', len(byte_string)) + byte_string cdef uint32_t length = len(byte_string)
return struct.pack('I', length) + byte_string
def from_bytes(self, data): def from_bytes(self, data):
self.vocab.serializer.unpack_into(data[4:], self) self.vocab.serializer.unpack_into(data[4:], self)