2020-07-09 17:39:31 +00:00
|
|
|
from typing import List
|
2023-06-14 15:48:41 +00:00
|
|
|
|
2020-07-09 17:39:31 +00:00
|
|
|
import numpy
|
2023-06-14 15:48:41 +00:00
|
|
|
import pytest
|
|
|
|
from numpy.testing import assert_array_almost_equal, assert_array_equal
|
|
|
|
from thinc.api import (
|
|
|
|
Adam,
|
|
|
|
Logistic,
|
|
|
|
Ragged,
|
|
|
|
Relu,
|
|
|
|
chain,
|
|
|
|
fix_random_seed,
|
|
|
|
reduce_mean,
|
|
|
|
set_dropout_rate,
|
|
|
|
)
|
|
|
|
|
2020-07-09 17:39:31 +00:00
|
|
|
from spacy.lang.en import English
|
|
|
|
from spacy.lang.en.examples import sentences as EN_SENTENCES
|
2023-06-14 15:48:41 +00:00
|
|
|
from spacy.ml.extract_spans import _get_span_indices, extract_spans
|
|
|
|
from spacy.ml.models import (
|
|
|
|
MaxoutWindowEncoder,
|
|
|
|
MultiHashEmbed,
|
|
|
|
build_bow_text_classifier,
|
|
|
|
build_simple_cnn_text_classifier,
|
|
|
|
build_spancat_model,
|
|
|
|
build_Tok2Vec_model,
|
|
|
|
)
|
|
|
|
from spacy.ml.staticvectors import StaticVectors
|
2023-12-21 10:00:06 +00:00
|
|
|
from spacy.util import registry
|
2020-07-09 17:39:31 +00:00
|
|
|
|
|
|
|
|
2020-10-18 12:50:41 +00:00
|
|
|
def get_textcat_bow_kwargs():
|
2020-07-28 20:43:19 +00:00
|
|
|
return {
|
2020-10-18 12:50:41 +00:00
|
|
|
"exclusive_classes": True,
|
2020-07-28 20:43:19 +00:00
|
|
|
"ngram_size": 1,
|
2020-10-18 12:50:41 +00:00
|
|
|
"no_output_layer": False,
|
|
|
|
"nO": 34,
|
2020-07-28 20:43:19 +00:00
|
|
|
}
|
|
|
|
|
2020-08-05 14:00:59 +00:00
|
|
|
|
2020-07-28 20:43:19 +00:00
|
|
|
def get_textcat_cnn_kwargs():
|
2022-10-25 12:53:18 +00:00
|
|
|
return {"tok2vec": make_test_tok2vec(), "exclusive_classes": False, "nO": 13}
|
2020-07-28 20:43:19 +00:00
|
|
|
|
2020-08-05 14:00:59 +00:00
|
|
|
|
2020-07-09 17:39:31 +00:00
|
|
|
def get_all_params(model):
|
|
|
|
params = []
|
|
|
|
for node in model.walk():
|
|
|
|
for name in node.param_names:
|
|
|
|
params.append(node.get_param(name).ravel())
|
|
|
|
return node.ops.xp.concatenate(params)
|
|
|
|
|
|
|
|
|
|
|
|
def get_docs():
|
|
|
|
nlp = English()
|
|
|
|
return list(nlp.pipe(EN_SENTENCES + [" ".join(EN_SENTENCES)]))
|
|
|
|
|
|
|
|
|
|
|
|
def get_gradient(model, Y):
|
|
|
|
if isinstance(Y, model.ops.xp.ndarray):
|
|
|
|
dY = model.ops.alloc(Y.shape, dtype=Y.dtype)
|
|
|
|
dY += model.ops.xp.random.uniform(-1.0, 1.0, Y.shape)
|
|
|
|
return dY
|
|
|
|
elif isinstance(Y, List):
|
|
|
|
return [get_gradient(model, y) for y in Y]
|
|
|
|
else:
|
2020-07-10 17:47:53 +00:00
|
|
|
raise ValueError(f"Could not get gradient for type {type(Y)}")
|
2020-07-09 17:39:31 +00:00
|
|
|
|
|
|
|
|
2020-07-28 20:43:19 +00:00
|
|
|
def get_tok2vec_kwargs():
|
|
|
|
# This actually creates models, so seems best to put it in a function.
|
|
|
|
return {
|
|
|
|
"embed": MultiHashEmbed(
|
2020-10-05 13:22:00 +00:00
|
|
|
width=32,
|
2020-10-05 17:57:45 +00:00
|
|
|
rows=[500, 500, 500],
|
2020-10-05 13:22:00 +00:00
|
|
|
attrs=["NORM", "PREFIX", "SHAPE"],
|
2020-10-05 19:58:18 +00:00
|
|
|
include_static_vectors=False,
|
2020-07-28 20:43:19 +00:00
|
|
|
),
|
|
|
|
"encode": MaxoutWindowEncoder(
|
2020-09-29 19:39:28 +00:00
|
|
|
width=32, depth=2, maxout_pieces=2, window_size=1
|
2020-08-05 14:00:59 +00:00
|
|
|
),
|
2020-07-28 20:43:19 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2022-10-25 12:53:18 +00:00
|
|
|
def make_test_tok2vec():
|
2020-07-28 20:43:19 +00:00
|
|
|
return build_Tok2Vec_model(**get_tok2vec_kwargs())
|
2020-07-09 17:39:31 +00:00
|
|
|
|
|
|
|
|
2020-10-05 13:22:00 +00:00
|
|
|
def test_multi_hash_embed():
|
|
|
|
embed = MultiHashEmbed(
|
|
|
|
width=32,
|
2020-10-05 17:57:45 +00:00
|
|
|
rows=[500, 500, 500],
|
2020-10-05 13:22:00 +00:00
|
|
|
attrs=["NORM", "PREFIX", "SHAPE"],
|
2020-10-05 19:58:18 +00:00
|
|
|
include_static_vectors=False,
|
2020-10-05 13:22:00 +00:00
|
|
|
)
|
|
|
|
hash_embeds = [node for node in embed.walk() if node.name == "hashembed"]
|
|
|
|
assert len(hash_embeds) == 3
|
|
|
|
# Check they look at different columns.
|
|
|
|
assert list(sorted(he.attrs["column"] for he in hash_embeds)) == [0, 1, 2]
|
|
|
|
# Check they use different seeds
|
|
|
|
assert len(set(he.attrs["seed"] for he in hash_embeds)) == 3
|
|
|
|
# Check they all have the same number of rows
|
|
|
|
assert [he.get_dim("nV") for he in hash_embeds] == [500, 500, 500]
|
|
|
|
# Now try with different row factors
|
|
|
|
embed = MultiHashEmbed(
|
|
|
|
width=32,
|
2020-10-05 17:57:45 +00:00
|
|
|
rows=[1000, 50, 250],
|
|
|
|
attrs=["NORM", "PREFIX", "SHAPE"],
|
2020-10-05 19:58:18 +00:00
|
|
|
include_static_vectors=False,
|
2020-10-05 13:22:00 +00:00
|
|
|
)
|
|
|
|
hash_embeds = [node for node in embed.walk() if node.name == "hashembed"]
|
|
|
|
assert [he.get_dim("nV") for he in hash_embeds] == [1000, 50, 250]
|
2020-10-05 19:58:18 +00:00
|
|
|
|
2020-10-05 13:22:00 +00:00
|
|
|
|
2020-07-09 17:39:31 +00:00
|
|
|
@pytest.mark.parametrize(
|
|
|
|
"seed,model_func,kwargs",
|
|
|
|
[
|
2020-07-28 20:43:19 +00:00
|
|
|
(0, build_Tok2Vec_model, get_tok2vec_kwargs()),
|
2020-10-18 12:50:41 +00:00
|
|
|
(0, build_bow_text_classifier, get_textcat_bow_kwargs()),
|
2020-07-28 20:43:19 +00:00
|
|
|
(0, build_simple_cnn_text_classifier, get_textcat_cnn_kwargs()),
|
2020-07-09 17:39:31 +00:00
|
|
|
],
|
|
|
|
)
|
|
|
|
def test_models_initialize_consistently(seed, model_func, kwargs):
|
|
|
|
fix_random_seed(seed)
|
|
|
|
model1 = model_func(**kwargs)
|
|
|
|
model1.initialize()
|
|
|
|
fix_random_seed(seed)
|
|
|
|
model2 = model_func(**kwargs)
|
|
|
|
model2.initialize()
|
|
|
|
params1 = get_all_params(model1)
|
|
|
|
params2 = get_all_params(model2)
|
2021-04-22 12:58:29 +00:00
|
|
|
assert_array_equal(model1.ops.to_numpy(params1), model2.ops.to_numpy(params2))
|
2020-07-09 17:39:31 +00:00
|
|
|
|
|
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
|
|
"seed,model_func,kwargs,get_X",
|
|
|
|
[
|
2020-07-28 20:43:19 +00:00
|
|
|
(0, build_Tok2Vec_model, get_tok2vec_kwargs(), get_docs),
|
2020-10-18 12:50:41 +00:00
|
|
|
(0, build_bow_text_classifier, get_textcat_bow_kwargs(), get_docs),
|
2020-07-28 20:43:19 +00:00
|
|
|
(0, build_simple_cnn_text_classifier, get_textcat_cnn_kwargs(), get_docs),
|
2020-07-09 17:39:31 +00:00
|
|
|
],
|
|
|
|
)
|
|
|
|
def test_models_predict_consistently(seed, model_func, kwargs, get_X):
|
|
|
|
fix_random_seed(seed)
|
|
|
|
model1 = model_func(**kwargs).initialize()
|
|
|
|
Y1 = model1.predict(get_X())
|
|
|
|
fix_random_seed(seed)
|
|
|
|
model2 = model_func(**kwargs).initialize()
|
|
|
|
Y2 = model2.predict(get_X())
|
|
|
|
|
|
|
|
if model1.has_ref("tok2vec"):
|
|
|
|
tok2vec1 = model1.get_ref("tok2vec").predict(get_X())
|
|
|
|
tok2vec2 = model2.get_ref("tok2vec").predict(get_X())
|
|
|
|
for i in range(len(tok2vec1)):
|
|
|
|
for j in range(len(tok2vec1[i])):
|
2020-07-22 11:42:59 +00:00
|
|
|
assert_array_equal(
|
2021-04-22 12:58:29 +00:00
|
|
|
numpy.asarray(model1.ops.to_numpy(tok2vec1[i][j])),
|
|
|
|
numpy.asarray(model2.ops.to_numpy(tok2vec2[i][j])),
|
2020-07-22 11:42:59 +00:00
|
|
|
)
|
2020-07-09 17:39:31 +00:00
|
|
|
|
2021-04-22 12:58:29 +00:00
|
|
|
try:
|
|
|
|
Y1 = model1.ops.to_numpy(Y1)
|
|
|
|
Y2 = model2.ops.to_numpy(Y2)
|
|
|
|
except Exception:
|
|
|
|
pass
|
2020-07-09 17:39:31 +00:00
|
|
|
if isinstance(Y1, numpy.ndarray):
|
|
|
|
assert_array_equal(Y1, Y2)
|
|
|
|
elif isinstance(Y1, List):
|
|
|
|
assert len(Y1) == len(Y2)
|
|
|
|
for y1, y2 in zip(Y1, Y2):
|
2021-04-22 12:58:29 +00:00
|
|
|
try:
|
|
|
|
y1 = model1.ops.to_numpy(y1)
|
|
|
|
y2 = model2.ops.to_numpy(y2)
|
|
|
|
except Exception:
|
|
|
|
pass
|
2020-07-09 17:39:31 +00:00
|
|
|
assert_array_equal(y1, y2)
|
|
|
|
else:
|
|
|
|
raise ValueError(f"Could not compare type {type(Y1)}")
|
|
|
|
|
|
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
|
|
"seed,dropout,model_func,kwargs,get_X",
|
|
|
|
[
|
2020-07-28 20:43:19 +00:00
|
|
|
(0, 0.2, build_Tok2Vec_model, get_tok2vec_kwargs(), get_docs),
|
2020-10-18 12:50:41 +00:00
|
|
|
(0, 0.2, build_bow_text_classifier, get_textcat_bow_kwargs(), get_docs),
|
2020-07-28 20:43:19 +00:00
|
|
|
(0, 0.2, build_simple_cnn_text_classifier, get_textcat_cnn_kwargs(), get_docs),
|
2020-07-09 17:39:31 +00:00
|
|
|
],
|
|
|
|
)
|
|
|
|
def test_models_update_consistently(seed, dropout, model_func, kwargs, get_X):
|
|
|
|
def get_updated_model():
|
|
|
|
fix_random_seed(seed)
|
|
|
|
optimizer = Adam(0.001)
|
|
|
|
model = model_func(**kwargs).initialize()
|
|
|
|
initial_params = get_all_params(model)
|
|
|
|
set_dropout_rate(model, dropout)
|
|
|
|
for _ in range(5):
|
|
|
|
Y, get_dX = model.begin_update(get_X())
|
|
|
|
dY = get_gradient(model, Y)
|
2020-07-22 11:42:59 +00:00
|
|
|
get_dX(dY)
|
2020-07-09 17:39:31 +00:00
|
|
|
model.finish_update(optimizer)
|
|
|
|
updated_params = get_all_params(model)
|
|
|
|
with pytest.raises(AssertionError):
|
2021-04-22 12:58:29 +00:00
|
|
|
assert_array_equal(
|
|
|
|
model.ops.to_numpy(initial_params), model.ops.to_numpy(updated_params)
|
|
|
|
)
|
2020-07-09 17:39:31 +00:00
|
|
|
return model
|
|
|
|
|
|
|
|
model1 = get_updated_model()
|
|
|
|
model2 = get_updated_model()
|
2021-04-22 12:58:29 +00:00
|
|
|
assert_array_almost_equal(
|
|
|
|
model1.ops.to_numpy(get_all_params(model1)),
|
|
|
|
model2.ops.to_numpy(get_all_params(model2)),
|
2021-10-13 08:47:56 +00:00
|
|
|
decimal=5,
|
2021-04-22 12:58:29 +00:00
|
|
|
)
|
2020-10-06 12:15:41 +00:00
|
|
|
|
|
|
|
|
2020-10-10 17:14:48 +00:00
|
|
|
@pytest.mark.parametrize("model_func,kwargs", [(StaticVectors, {"nO": 128, "nM": 300})])
|
2020-10-06 12:15:41 +00:00
|
|
|
def test_empty_docs(model_func, kwargs):
|
|
|
|
nlp = English()
|
|
|
|
model = model_func(**kwargs).initialize()
|
|
|
|
# Test the layer can be called successfully with 0, 1 and 2 empty docs.
|
|
|
|
for n_docs in range(3):
|
|
|
|
docs = [nlp("") for _ in range(n_docs)]
|
|
|
|
# Test predict
|
2020-10-10 17:14:48 +00:00
|
|
|
model.predict(docs)
|
2020-10-06 12:15:41 +00:00
|
|
|
# Test backprop
|
|
|
|
output, backprop = model.begin_update(docs)
|
2020-10-10 17:14:48 +00:00
|
|
|
backprop(output)
|
2021-06-24 10:35:27 +00:00
|
|
|
|
|
|
|
|
|
|
|
def test_init_extract_spans():
|
2021-07-18 05:44:56 +00:00
|
|
|
extract_spans().initialize()
|
2021-06-24 10:35:27 +00:00
|
|
|
|
|
|
|
|
|
|
|
def test_extract_spans_span_indices():
|
|
|
|
model = extract_spans().initialize()
|
|
|
|
spans = Ragged(
|
|
|
|
model.ops.asarray([[0, 3], [2, 3], [5, 7]], dtype="i"),
|
|
|
|
model.ops.asarray([2, 1], dtype="i"),
|
|
|
|
)
|
|
|
|
x_lengths = model.ops.asarray([5, 10], dtype="i")
|
|
|
|
indices = _get_span_indices(model.ops, spans, x_lengths)
|
|
|
|
assert list(indices) == [0, 1, 2, 2, 10, 11]
|
|
|
|
|
|
|
|
|
|
|
|
def test_extract_spans_forward_backward():
|
|
|
|
model = extract_spans().initialize()
|
|
|
|
X = Ragged(model.ops.alloc2f(15, 4), model.ops.asarray([5, 10], dtype="i"))
|
|
|
|
spans = Ragged(
|
|
|
|
model.ops.asarray([[0, 3], [2, 3], [5, 7]], dtype="i"),
|
|
|
|
model.ops.asarray([2, 1], dtype="i"),
|
|
|
|
)
|
|
|
|
Y, backprop = model.begin_update((X, spans))
|
|
|
|
assert list(Y.lengths) == [3, 1, 2]
|
|
|
|
assert Y.dataXd.shape == (6, 4)
|
|
|
|
dX, spans2 = backprop(Y)
|
|
|
|
assert spans2 is spans
|
|
|
|
assert dX.dataXd.shape == X.dataXd.shape
|
|
|
|
assert list(dX.lengths) == list(X.lengths)
|
|
|
|
|
|
|
|
|
|
|
|
def test_spancat_model_init():
|
|
|
|
model = build_spancat_model(
|
|
|
|
build_Tok2Vec_model(**get_tok2vec_kwargs()), reduce_mean(), Logistic()
|
|
|
|
)
|
|
|
|
model.initialize()
|
|
|
|
|
|
|
|
|
|
|
|
def test_spancat_model_forward_backward(nO=5):
|
|
|
|
tok2vec = build_Tok2Vec_model(**get_tok2vec_kwargs())
|
|
|
|
docs = get_docs()
|
|
|
|
spans_list = []
|
|
|
|
lengths = []
|
|
|
|
for doc in docs:
|
|
|
|
spans_list.append(doc[:2])
|
|
|
|
spans_list.append(doc[1:4])
|
|
|
|
lengths.append(2)
|
|
|
|
spans = Ragged(
|
|
|
|
tok2vec.ops.asarray([[s.start, s.end] for s in spans_list], dtype="i"),
|
|
|
|
tok2vec.ops.asarray(lengths, dtype="i"),
|
|
|
|
)
|
|
|
|
model = build_spancat_model(
|
|
|
|
tok2vec, reduce_mean(), chain(Relu(nO=nO), Logistic())
|
|
|
|
).initialize(X=(docs, spans))
|
|
|
|
|
|
|
|
Y, backprop = model((docs, spans), is_train=True)
|
|
|
|
assert Y.shape == (spans.dataXd.shape[0], nO)
|
|
|
|
backprop(Y)
|
2023-12-21 10:00:06 +00:00
|
|
|
|
|
|
|
|
|
|
|
def test_textcat_reduce_invalid_args():
|
|
|
|
textcat_reduce = registry.architectures.get("spacy.TextCatReduce.v1")
|
|
|
|
tok2vec = make_test_tok2vec()
|
|
|
|
with pytest.raises(ValueError, match=r"must be used with at least one reduction"):
|
|
|
|
textcat_reduce(
|
|
|
|
tok2vec=tok2vec,
|
|
|
|
exclusive_classes=False,
|
|
|
|
use_reduce_first=False,
|
|
|
|
use_reduce_last=False,
|
|
|
|
use_reduce_max=False,
|
|
|
|
use_reduce_mean=False,
|
|
|
|
)
|