mirror of https://github.com/explosion/spaCy.git
26 lines
833 B
Python
26 lines
833 B
Python
|
# coding: utf-8
|
||
|
|
from __future__ import unicode_literals
|
||
|
|
|
||
|
|
from ..util import make_tempdir
|
||
|
|
from ...pipeline import TokenVectorEncoder as Tensorizer
|
||
|
|
|
||
|
|
import pytest
|
||
|
|
|
||
|
|
|
||
|
|
def test_serialize_tensorizer_roundtrip_bytes(en_vocab):
|
||
|
|
tensorizer = Tensorizer(en_vocab)
|
||
|
|
tensorizer.model = tensorizer.Model()
|
||
|
|
tensorizer_b = tensorizer.to_bytes()
|
||
|
|
new_tensorizer = Tensorizer(en_vocab).from_bytes(tensorizer_b)
|
||
|
|
assert new_tensorizer.to_bytes() == tensorizer_b
|
||
|
|
|
||
|
|
|
||
|
|
def test_serialize_tensorizer_roundtrip_disk(en_vocab):
|
||
|
|
tensorizer = Tensorizer(en_vocab)
|
||
|
|
tensorizer.model = tensorizer.Model()
|
||
|
|
with make_tempdir() as d:
|
||
|
|
file_path = d / 'tensorizer'
|
||
|
|
tensorizer.to_disk(file_path)
|
||
|
|
tensorizer_d = Tensorizer(en_vocab).from_disk(file_path)
|
||
|
|
assert tensorizer.to_bytes() == tensorizer_d.to_bytes()
|