explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-03-12 07:15:48 +03:00
Code Issues Packages Projects Releases Wiki Activity

Update tests

Browse Source
This commit is contained in:
Matthew Honnibal 2020-06-22 14:26:38 +02:00
parent 6a75992af6
commit 792bf9476c
1 changed files with 7 additions and 47 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

54
spacy/tests/test_gold.py
View File

@ -349,6 +349,8 @@ def test_iob_to_biluo():
iob_to_biluo(bad_iob)
# This test is outdated as we use DocBin now. It should probably be removed?
@pytest.mark.xfail(reason="Outdated")
def test_roundtrip_docs_to_json(doc):
nlp = English()
text = doc.text
@ -366,7 +368,7 @@ def test_roundtrip_docs_to_json(doc):
with make_tempdir() as tmpdir:
json_file = tmpdir / "roundtrip.json"
srsly.write_json(json_file, [docs_to_json(doc)])
goldcorpus = Corpus(train=str(json_file), dev=str(json_file))
goldcorpus = Corpus(str(json_file), str(json_file))
reloaded_example = next(goldcorpus.dev_dataset(nlp=nlp))
assert len(doc) == goldcorpus.count_train()
@ -387,39 +389,10 @@ def test_roundtrip_docs_to_json(doc):
assert cats["BAKING"] == reloaded_example.reference.cats["BAKING"]
@pytest.mark.xfail # TODO do we need to do the projectivity differently?
def test_projective_train_vs_nonprojective_dev(doc):
nlp = English()
deps = [t.dep_ for t in doc]
heads = [t.head.i for t in doc]
with make_tempdir() as tmpdir:
json_file = tmpdir / "test.json"
# write to JSON train dicts
srsly.write_json(json_file, [docs_to_json(doc)])
goldcorpus = Corpus(str(json_file), str(json_file))
train_reloaded_example = next(goldcorpus.train_dataset(nlp))
train_goldparse = get_parses_from_example(train_reloaded_example)[0][1]
dev_reloaded_example = next(goldcorpus.dev_dataset(nlp))
dev_goldparse = get_parses_from_example(dev_reloaded_example)[0][1]
assert is_nonproj_tree([t.head.i for t in doc]) is True
assert is_nonproj_tree(train_goldparse.heads) is False
assert heads[:-1] == train_goldparse.heads[:-1]
assert heads[-1] != train_goldparse.heads[-1]
assert deps[:-1] == train_goldparse.labels[:-1]
assert deps[-1] != train_goldparse.labels[-1]
assert heads == dev_goldparse.heads
assert deps == dev_goldparse.labels
# Hm, not sure where misalignment check would be handled? In the components too?
# I guess that does make sense. A text categorizer doesn't care if it's
# misaligned...
@pytest.mark.xfail # TODO
@pytest.mark.xfail(reason="Outdated")
def test_ignore_misaligned(doc):
nlp = English()
text = doc.text
@ -450,6 +423,9 @@ def test_ignore_misaligned(doc):
assert len(train_reloaded_example) == 0
# We probably want the orth variant logic back, but this test won't be quite
# right -- we need to go from DocBin.
@pytest.mark.xfail(reason="Outdated")
def test_make_orth_variants(doc):
nlp = English()
with make_tempdir() as tmpdir:
@ -594,19 +570,3 @@ def test_split_sents(merged_dict):
assert token_annotation_2["words"] == ["It", "is", "just", "me"]
assert token_annotation_2["tags"] == ["PRON", "AUX", "ADV", "PRON"]
assert token_annotation_2["sent_starts"] == [1, 0, 0, 0]
# This fails on some None value? Need to look into that.
@pytest.mark.xfail # TODO
def test_tuples_to_example(vocab, merged_dict):
cats = {"TRAVEL": 1.0, "BAKING": 0.0}
merged_dict = dict(merged_dict)
merged_dict["cats"] = cats
ex = Example.from_dict(Doc(vocab, words=merged_dict["words"]), merged_dict)
words = [token.text for token in ex.reference]
assert words == merged_dict["words"]
tags = [token.tag_ for token in ex.reference]
assert tags == merged_dict["tags"]
sent_starts = [token.is_sent_start for token in ex.reference]
assert sent_starts == [bool(v) for v in merged_dict["sent_starts"]]
ex.reference.cats == cats