explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-03-13 07:55:49 +03:00
Code Issues Packages Projects Releases Wiki Activity

Update gold test for NER

Browse Source
This commit is contained in:
Matthew Honnibal 2020-06-24 15:19:40 +02:00
parent 5120113f39
commit a78427d640
1 changed files with 22 additions and 53 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

75
spacy/tests/test_gold.py
View File

@ -204,9 +204,9 @@ def test_gold_biluo_different_tokenization(en_vocab, en_tokenizer):
entities = [(len("I flew to "), len("I flew to San Francisco Valley"), "LOC")] entities = [(len("I flew to "), len("I flew to San Francisco Valley"), "LOC")]
gold_words = ["I", "flew", "to", "San", "Francisco", "Valley", "."] gold_words = ["I", "flew", "to", "San", "Francisco", "Valley", "."]
example = Example.from_dict(doc, {"words": gold_words, "entities": entities}) example = Example.from_dict(doc, {"words": gold_words, "entities": entities})
assert example.get_aligned("ENT_IOB") == [2, 2, 3, 2] ner_tags = example.get_aligned_ner()
assert example.get_aligned("ENT_TYPE", as_string=True) == ["", "", "LOC", ""] assert ner_tags == ["O", "O", "U-LOC", "O"]
# many-to-one # many-to-one
words = ["I", "flew", "to", "San", "Francisco", "Valley", "."] words = ["I", "flew", "to", "San", "Francisco", "Valley", "."]
spaces = [True, True, True, True, True, False, False] spaces = [True, True, True, True, True, False, False]
@ -214,16 +214,8 @@ def test_gold_biluo_different_tokenization(en_vocab, en_tokenizer):
entities = [(len("I flew to "), len("I flew to San Francisco Valley"), "LOC")] entities = [(len("I flew to "), len("I flew to San Francisco Valley"), "LOC")]
gold_words = ["I", "flew to", "San Francisco Valley", "."] gold_words = ["I", "flew to", "San Francisco Valley", "."]
example = Example.from_dict(doc, {"words": gold_words, "entities": entities}) example = Example.from_dict(doc, {"words": gold_words, "entities": entities})
assert example.get_aligned("ENT_IOB") == [2, 2, 2, 3, 1, 1, 2] ner_tags = example.get_aligned_ner()
assert example.get_aligned("ENT_TYPE", as_string=True) == [ assert ner_tags == ["O", "O", "O", "B-LOC", "I-LOC", "L-LOC", "O"]
"",
"",
"",
"LOC",
"LOC",
"LOC",
"",
]
# misaligned # misaligned
words = ["I flew", "to", "San Francisco", "Valley", "."] words = ["I flew", "to", "San Francisco", "Valley", "."]
@ -237,15 +229,15 @@ def test_gold_biluo_different_tokenization(en_vocab, en_tokenizer):
example = Example.from_dict( example = Example.from_dict(
doc, {"words": gold_words, "entities": entities, "links": links} doc, {"words": gold_words, "entities": entities, "links": links}
) )
assert example.get_aligned("ENT_IOB") == [2, 2, 3, 1, 2] ner_tags = example.get_aligned_ner()
assert example.get_aligned("ENT_TYPE", as_string=True) == ["", "", "LOC", "LOC", ""] assert ner_tags == ["O", "O", "B-LOC", "L-LOC", "O"]
assert example.get_aligned("ENT_KB_ID", as_string=True) == [ #assert example.get_aligned("ENT_KB_ID", as_string=True) == [
"", # "",
"", # "",
"Q816843", # "Q816843",
"Q816843", # "Q816843",
"", # "",
] #]
assert example.to_dict()["doc_annotation"]["links"][(offset_start, offset_end)] == { assert example.to_dict()["doc_annotation"]["links"][(offset_start, offset_end)] == {
"Q816843": 1.0 "Q816843": 1.0
} }
@ -262,16 +254,8 @@ def test_gold_biluo_different_tokenization(en_vocab, en_tokenizer):
example = Example.from_dict( example = Example.from_dict(
doc, {"words": gold_words, "spaces": gold_spaces, "entities": entities} doc, {"words": gold_words, "spaces": gold_spaces, "entities": entities}
) )
assert example.get_aligned("ENT_IOB") == [2, 2, 2, 2, 3, 1, 2] ner_tags = example.get_aligned_ner()
assert example.get_aligned("ENT_TYPE", as_string=True) == [ assert ner_tags == ["O", "O", "O", "O", "B-LOC", "L-LOC", "O"]
"",
"",
"",
"",
"LOC",
"LOC",
"",
]
# from issue #4791 # from issue #4791
doc = en_tokenizer("I'll return the ₹54 amount") doc = en_tokenizer("I'll return the ₹54 amount")
@ -281,15 +265,8 @@ def test_gold_biluo_different_tokenization(en_vocab, en_tokenizer):
example = Example.from_dict( example = Example.from_dict(
doc, {"words": gold_words, "spaces": gold_spaces, "entities": entities} doc, {"words": gold_words, "spaces": gold_spaces, "entities": entities}
) )
assert example.get_aligned("ENT_IOB") == [2, 2, 2, 2, 3, 2] ner_tags = example.get_aligned_ner()
assert example.get_aligned("ENT_TYPE", as_string=True) == [ assert ner_tags == ["O", "O", "O", "O", "U-MONEY", "O"]
"",
"",
"",
"",
"MONEY",
"",
]
doc = en_tokenizer("I'll return the $54 amount") doc = en_tokenizer("I'll return the $54 amount")
gold_words = ["I", "'ll", "return", "the", "$", "54", "amount"] gold_words = ["I", "'ll", "return", "the", "$", "54", "amount"]
@ -298,16 +275,8 @@ def test_gold_biluo_different_tokenization(en_vocab, en_tokenizer):
example = Example.from_dict( example = Example.from_dict(
doc, {"words": gold_words, "spaces": gold_spaces, "entities": entities} doc, {"words": gold_words, "spaces": gold_spaces, "entities": entities}
) )
assert example.get_aligned("ENT_IOB") == [2, 2, 2, 2, 3, 1, 2] ner_tags = example.get_aligned_ner()
assert example.get_aligned("ENT_TYPE", as_string=True) == [ assert ner_tags == ["O", "O", "O", "O", "B-MONEY", "L-MONEY", "O"]
"",
"",
"",
"",
"MONEY",
"MONEY",
"",
]
def test_roundtrip_offsets_biluo_conversion(en_tokenizer): def test_roundtrip_offsets_biluo_conversion(en_tokenizer):
@ -484,8 +453,8 @@ def test_goldparse_startswith_space(en_tokenizer):
example = Example.from_dict( example = Example.from_dict(
doc, {"words": gold_words, "entities": entities, "deps": deps, "heads": heads} doc, {"words": gold_words, "entities": entities, "deps": deps, "heads": heads}
) )
assert example.get_aligned("ENT_IOB") == [None, 3] ner_tags = example.get_aligned_ner()
assert example.get_aligned("ENT_TYPE", as_string=True) == [None, "DATE"] assert ner_tags == ["O", "U-DATE"]
assert example.get_aligned("DEP", as_string=True) == [None, "ROOT"] assert example.get_aligned("DEP", as_string=True) == [None, "ROOT"]