Skip to content
Snippets Groups Projects
Commit d77e9801 authored by Blanche Miret's avatar Blanche Miret Committed by kermorvant
Browse files

Handle all beginning labels

parent 3b8f584e
No related branches found
No related tags found
1 merge request!8Handle all beginning labels
Hide whitespace changes
Inline Side-by-side
nerval/evaluate.py
+ 36
− 11
View file @ d77e9801
...@@ -12,6 +12,7 @@ import termtables as tt ...@@ -12,6 +12,7 @@ import termtables as tt
NOT_ENTITY_TAG = "O" NOT_ENTITY_TAG = "O"
THRESHOLD = 0.30 THRESHOLD = 0.30
BEGINNING_POS = ["B", "S", "U"]
def get_type_label(label: str) -> str: def get_type_label(label: str) -> str:
...@@ -31,6 +32,23 @@ def get_type_label(label: str) -> str: ...@@ -31,6 +32,23 @@ def get_type_label(label: str) -> str:
return tag return tag
def get_position_label(label: str) -> str:
"""Return the position of a label
Input format: "[BIELUS]-type"
"""
try:
pos = (
NOT_ENTITY_TAG
if label == NOT_ENTITY_TAG
else re.match(r"([BIELUS])-.{3,4}", label)[1]
)
except TypeError:
raise (Exception(f"The label {label} is not valid in BIOES/BILOU format."))
return pos
def parse_bio(path: str) -> dict: def parse_bio(path: str) -> dict:
"""Parse a BIO file to get text content, character-level NE labels and entity types count. """Parse a BIO file to get text content, character-level NE labels and entity types count.
...@@ -75,13 +93,17 @@ def parse_bio(path: str) -> dict: ...@@ -75,13 +93,17 @@ def parse_bio(path: str) -> dict:
if index != 0: if index != 0:
# If new word has same tag as previous, not new entity and in entity, continue entity # If new word has same tag as previous, not new entity and in entity, continue entity
if last_tag == tag and "B" not in label and tag != NOT_ENTITY_TAG: if (
last_tag == tag
and get_position_label(label) not in BEGINNING_POS
and tag != NOT_ENTITY_TAG
):
labels.append(f"I-{last_tag}") labels.append(f"I-{last_tag}")
# If new word begins a new entity of different type, check for nested entity to correctly tag the space # If new word begins a new entity of different type, check for nested entity to correctly tag the space
elif ( elif (
last_tag != tag last_tag != tag
and "B" in label and get_position_label(label) in BEGINNING_POS
and tag != NOT_ENTITY_TAG and tag != NOT_ENTITY_TAG
and last_tag != NOT_ENTITY_TAG and last_tag != NOT_ENTITY_TAG
): ):
...@@ -99,7 +121,7 @@ def parse_bio(path: str) -> dict: ...@@ -99,7 +121,7 @@ def parse_bio(path: str) -> dict:
# Check for continuation of the original entity # Check for continuation of the original entity
if ( if (
index < len(lines) index < len(lines)
and "B" not in future_label and get_position_label(future_label) not in BEGINNING_POS
and get_type_label(future_label) == last_tag and get_type_label(future_label) == last_tag
): ):
labels.append(f"I-{last_tag}") labels.append(f"I-{last_tag}")
...@@ -117,13 +139,13 @@ def parse_bio(path: str) -> dict: ...@@ -117,13 +139,13 @@ def parse_bio(path: str) -> dict:
in_nested_entity = False in_nested_entity = False
# Add a tag for each letter in the word # Add a tag for each letter in the word
if "B" in label: if get_position_label(label) in BEGINNING_POS:
labels += [f"B-{tag}"] + [f"I-{tag}"] * (len(word) - 1) labels += [f"B-{tag}"] + [f"I-{tag}"] * (len(word) - 1)
else: else:
labels += [label] * len(word) labels += [label] * len(word)
# Count nb entity for each type # Count nb entity for each type
if "B" in label: if get_position_label(label) in BEGINNING_POS:
entity_count[tag] = entity_count.get(tag, 0) + 1 entity_count[tag] = entity_count.get(tag, 0) + 1
entity_count["All"] += 1 entity_count["All"] += 1
...@@ -171,7 +193,7 @@ def look_for_further_entity_part(index, tag, characters, labels): ...@@ -171,7 +193,7 @@ def look_for_further_entity_part(index, tag, characters, labels):
index += 1 index += 1
while ( while (
index < len(characters) index < len(characters)
and "B" not in labels[index] and get_position_label(labels[index]) not in BEGINNING_POS
and get_type_label(labels[index]) == tag and get_type_label(labels[index]) == tag
): ):
visited.append(index) visited.append(index)
...@@ -254,7 +276,7 @@ def compute_matches( ...@@ -254,7 +276,7 @@ def compute_matches(
else: else:
# If beginning new entity # If beginning new entity
if "B" in label_ref: if get_position_label(label_ref) in BEGINNING_POS:
current_ref, current_compar = [], [] current_ref, current_compar = [], []
last_tag = tag_ref last_tag = tag_ref
found_aligned_beginning = False found_aligned_beginning = False
...@@ -269,18 +291,21 @@ def compute_matches( ...@@ -269,18 +291,21 @@ def compute_matches(
continue continue
# If just beginning new entity, backtrack tags on prediction string # If just beginning new entity, backtrack tags on prediction string
if len(current_ref) == 1 and "B" not in labels_predict[i]: if (
len(current_ref) == 1
and get_position_label(labels_predict[i]) not in BEGINNING_POS
):
j = i - 1 j = i - 1
while ( while (
j >= 0 j >= 0
and get_type_label(labels_predict[j]) == tag_ref and get_type_label(labels_predict[j]) == tag_ref
and "B" not in labels_predict[j] and get_position_label(labels_predict[j]) not in BEGINNING_POS
and j not in visited_predict and j not in visited_predict
): ):
j -= 1 j -= 1
if ( if (
"B" in labels_predict[j] get_position_label(labels_predict[j]) in BEGINNING_POS
and get_type_label(labels_predict[j]) == tag_ref and get_type_label(labels_predict[j]) == tag_ref
and j not in visited_predict and j not in visited_predict
): ):
...@@ -372,7 +397,7 @@ def get_labels_aligned(original: str, aligned: str, labels_original: list) -> li ...@@ -372,7 +397,7 @@ def get_labels_aligned(original: str, aligned: str, labels_original: list) -> li
elif not char == original[index_original]: elif not char == original[index_original]:
new_label = ( new_label = (
last_label last_label
if "B" not in last_label if get_position_label(last_label) not in BEGINNING_POS
else f"I-{get_type_label(last_label)}" else f"I-{get_type_label(last_label)}"
) )
......
tests/bioues.bio 0 → 100644
+ 10
− 0
View file @ d77e9801
Gérard B-PER
de I-PER
Nerval I-PER
was O
born O
in O
Paris U-LOC
in O
1808 S-DAT
. O
tests/test_parse_bio.py
+ 97
− 44
View file @ d77e9801
...@@ -8,58 +8,110 @@ EMPTY_BIO = "tests/test_empty.bio" ...@@ -8,58 +8,110 @@ EMPTY_BIO = "tests/test_empty.bio"
BAD_BIO = "tests/test_bad.bio" BAD_BIO = "tests/test_bad.bio"
FAKE_ANNOT_BIO = "tests/test_annot.bio" FAKE_ANNOT_BIO = "tests/test_annot.bio"
FAKE_PREDICT_BIO = "tests/test_predict.bio" FAKE_PREDICT_BIO = "tests/test_predict.bio"
BIOUES_BIO = "tests/bioues.bio"
# fmt: off
expected_parsed_annot = { expected_parsed_annot = {
'entity_count': {'All': 3, 'DAT': 1, 'LOC': 1, 'PER': 1}, "entity_count": {"All": 3, "DAT": 1, "LOC": 1, "PER": 1},
'labels': [ "labels": [
'B-PER', 'I-PER', 'I-PER', 'I-PER', 'I-PER', 'I-PER', "B-PER",
'I-PER', "I-PER",
'I-PER', 'I-PER', "I-PER",
'I-PER', "I-PER",
'I-PER', 'I-PER', 'I-PER', 'I-PER', 'I-PER', 'I-PER', "I-PER",
'O', "I-PER",
'O', 'O', 'O', "I-PER",
'O', "I-PER",
'O', 'O', 'O', 'O', "I-PER",
'O', "I-PER",
'O', 'O', "I-PER",
'O', "I-PER",
'B-LOC', 'I-LOC', 'I-LOC', 'I-LOC', 'I-LOC', "I-PER",
'O', "I-PER",
'O', 'O', "I-PER",
'O', "I-PER",
'B-DAT', 'I-DAT', 'I-DAT', 'I-DAT', "O",
'O', "O",
'O' "O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"B-LOC",
"I-LOC",
"I-LOC",
"I-LOC",
"I-LOC",
"O",
"O",
"O",
"O",
"B-DAT",
"I-DAT",
"I-DAT",
"I-DAT",
"O",
"O",
], ],
'words': 'Gérard de Nerval was born in Paris in 1808 .' "words": "Gérard de Nerval was born in Paris in 1808 .",
} }
expected_parsed_predict = { expected_parsed_predict = {
'entity_count': {'All': 3, 'DAT': 1, '***': 1, 'PER': 1}, "entity_count": {"All": 3, "DAT": 1, "***": 1, "PER": 1},
'labels': [ "labels": [
'B-PER', 'I-PER', 'I-PER', 'I-PER', 'I-PER', 'I-PER', "B-PER",
'I-PER', "I-PER",
'I-PER', 'I-PER', "I-PER",
'I-PER', "I-PER",
'I-PER', 'I-PER', 'I-PER', 'I-PER', 'I-PER', 'I-PER', 'I-PER', 'I-PER', "I-PER",
'O', "I-PER",
'O', 'O', 'O', 'O', 'O', "I-PER",
'O', "I-PER",
'O', 'O', "I-PER",
'O', "I-PER",
'B-***', 'I-***', 'I-***', 'I-***', 'I-***', "I-PER",
'O', "I-PER",
'O', 'O', "I-PER",
'O', "I-PER",
'B-DAT', 'I-DAT', 'I-DAT', 'I-DAT', "I-PER",
'O', "I-PER",
'O', 'O' "I-PER",
"I-PER",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"B-***",
"I-***",
"I-***",
"I-***",
"I-***",
"O",
"O",
"O",
"O",
"B-DAT",
"I-DAT",
"I-DAT",
"I-DAT",
"O",
"O",
"O",
], ],
'words': 'G*rard de *N*erval bo*rn in Paris in 1833 *.' "words": "G*rard de *N*erval bo*rn in Paris in 1833 *.",
} }
# fmt: on
@pytest.mark.parametrize( @pytest.mark.parametrize(
...@@ -68,6 +120,7 @@ expected_parsed_predict = { ...@@ -68,6 +120,7 @@ expected_parsed_predict = {
(FAKE_ANNOT_BIO, expected_parsed_annot), (FAKE_ANNOT_BIO, expected_parsed_annot),
(FAKE_PREDICT_BIO, expected_parsed_predict), (FAKE_PREDICT_BIO, expected_parsed_predict),
(EMPTY_BIO, None), (EMPTY_BIO, None),
(BIOUES_BIO, expected_parsed_annot),
], ],
) )
def test_parse_bio(test_input, expected): def test_parse_bio(test_input, expected):
......
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment