spaCy/spacy/tests/pipeline/test_models.py

108 lines
3.7 KiB
Python
Raw Normal View History

from typing import List
2020-10-13 19:50:07 +03:00
import numpy
import pytest
2020-10-13 19:50:07 +03:00
from numpy.testing import assert_almost_equal
from spacy.vocab import Vocab
from thinc.api import NumpyOps, Model, data_validation
2020-10-13 19:50:07 +03:00
from thinc.types import Array2d, Ragged
from spacy.lang.en import English
2020-10-13 19:50:07 +03:00
from spacy.ml import FeatureExtractor, StaticVectors
from spacy.ml._character_embed import CharacterEmbed
from spacy.tokens import Doc
OPS = NumpyOps()
2020-10-13 19:50:07 +03:00
texts = ["These are 4 words", "Here just three"]
l0 = [[1, 2], [3, 4], [5, 6], [7, 8]]
l1 = [[9, 8], [7, 6], [5, 4]]
2020-10-13 19:50:07 +03:00
list_floats = [OPS.xp.asarray(l0, dtype="f"), OPS.xp.asarray(l1, dtype="f")]
list_ints = [OPS.xp.asarray(l0, dtype="i"), OPS.xp.asarray(l1, dtype="i")]
array = OPS.xp.asarray(l1, dtype="f")
ragged = Ragged(array, OPS.xp.asarray([2, 1], dtype="i"))
def get_docs():
vocab = Vocab()
for t in texts:
for word in t.split():
hash_id = vocab.strings.add(word)
vector = numpy.random.uniform(-1, 1, (7,))
vocab.set_vector(hash_id, vector)
docs = [English(vocab)(t) for t in texts]
return docs
# Test components with a model of type Model[List[Doc], List[Floats2d]]
@pytest.mark.parametrize("name", ["tagger", "tok2vec", "morphologizer", "senter"])
2020-10-13 19:50:07 +03:00
def test_components_batching_list(name):
nlp = English()
proc = nlp.create_pipe(name)
2020-10-13 19:52:37 +03:00
util_batch_unbatch_docs_list(proc.model, get_docs(), list_floats)
# Test components with a model of type Model[List[Doc], Floats2d]
@pytest.mark.parametrize("name", ["textcat"])
2020-10-13 19:50:07 +03:00
def test_components_batching_array(name):
nlp = English()
proc = nlp.create_pipe(name)
2020-10-13 19:52:37 +03:00
util_batch_unbatch_docs_array(proc.model, get_docs(), array)
2020-10-13 19:50:07 +03:00
LAYERS = [
(CharacterEmbed(nM=5, nC=3), get_docs(), list_floats),
(FeatureExtractor([100, 200]), get_docs(), list_ints),
(StaticVectors(), get_docs(), ragged),
]
@pytest.mark.parametrize("model,in_data,out_data", LAYERS)
def test_layers_batching_all(model, in_data, out_data):
# In = List[Doc]
if isinstance(in_data, list) and isinstance(in_data[0], Doc):
if isinstance(out_data, OPS.xp.ndarray) and out_data.ndim == 2:
2020-10-13 19:52:37 +03:00
util_batch_unbatch_docs_array(model, in_data, out_data)
2020-10-13 19:50:07 +03:00
elif (
isinstance(out_data, list)
and isinstance(out_data[0], OPS.xp.ndarray)
and out_data[0].ndim == 2
):
2020-10-13 19:52:37 +03:00
util_batch_unbatch_docs_list(model, in_data, out_data)
2020-10-13 19:50:07 +03:00
elif isinstance(out_data, Ragged):
2020-10-13 19:52:37 +03:00
util_batch_unbatch_docs_ragged(model, in_data, out_data)
2020-10-13 19:50:07 +03:00
2020-10-13 19:52:37 +03:00
def util_batch_unbatch_docs_list(
2020-10-13 19:50:07 +03:00
model: Model[List[Doc], List[Array2d]], in_data: List[Doc], out_data: List[Array2d]
):
with data_validation(True):
model.initialize(in_data, out_data)
Y_batched = model.predict(in_data)
Y_not_batched = [model.predict([u])[0] for u in in_data]
2020-10-13 19:50:07 +03:00
for i in range(len(Y_batched)):
assert_almost_equal(Y_batched[i], Y_not_batched[i], decimal=4)
2020-10-13 19:52:37 +03:00
def util_batch_unbatch_docs_array(
2020-10-13 19:50:07 +03:00
model: Model[List[Doc], Array2d], in_data: List[Doc], out_data: Array2d
):
with data_validation(True):
model.initialize(in_data, out_data)
Y_batched = model.predict(in_data).tolist()
Y_not_batched = [model.predict([u])[0] for u in in_data]
assert_almost_equal(Y_batched, Y_not_batched, decimal=4)
2020-10-13 19:52:37 +03:00
def util_batch_unbatch_docs_ragged(
2020-10-13 19:50:07 +03:00
model: Model[List[Doc], Ragged], in_data: List[Doc], out_data: Ragged
):
with data_validation(True):
model.initialize(in_data, out_data)
Y_batched = model.predict(in_data)
Y_not_batched = []
for u in in_data:
Y_not_batched.extend(model.predict([u]).data.tolist())
assert_almost_equal(Y_batched.data, Y_not_batched, decimal=4)