Spaces:
Running
Running
File size: 5,538 Bytes
3891395 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 |
"""
Evaluation for two strings or list of strings.
Code taken from the DROP benchmark - https://github.com/allenai/allennlp-reading-comprehension/blob/master/allennlp_rc/eval/drop_eval.py
"""
from collections import defaultdict
from typing import List, Set, Tuple, Union
import string
import re
import numpy as np
from scipy.optimize import linear_sum_assignment
# From here through _normalize_answer was originally copied from:
# https://worksheets.codalab.org/rest/bundles/0x6b567e1cf2e041ec80d7098f031c5c9e/contents/blob/
# Then cleaned up and modified a bit.
def _remove_articles(text: str) -> str:
regex = re.compile(r"\b(a|an|the)\b", re.UNICODE)
return re.sub(regex, " ", text)
def _white_space_fix(text: str) -> str:
return " ".join(text.split())
EXCLUDE = set(string.punctuation)
def _remove_punc(text: str) -> str:
if not _is_number(text):
return "".join(ch for ch in text if ch not in EXCLUDE)
else:
return text
def _lower(text: str) -> str:
return text.lower()
def _tokenize(text: str) -> List[str]:
return re.split(" |-", text)
def _normalize_answer(text: str) -> str:
"""Lower text and remove punctuation, articles and extra whitespace."""
parts = [
_white_space_fix(
_remove_articles(_normalize_number(_remove_punc(_lower(token))))
)
for token in _tokenize(text)
]
parts = [part for part in parts if part.strip()]
normalized = " ".join(parts).strip()
return normalized
def _is_number(text: str) -> bool:
try:
float(text)
return True
except ValueError:
return False
def _normalize_number(text: str) -> str:
if _is_number(text):
return str(float(text))
else:
return text
def _answer_to_bags(
answer: Union[str, List[str], Tuple[str, ...]]
) -> Tuple[List[str], List[Set[str]]]:
if isinstance(answer, (list, tuple)):
raw_spans = answer
else:
raw_spans = [answer]
normalized_spans: List[str] = []
token_bags = []
for raw_span in raw_spans:
normalized_span = _normalize_answer(raw_span)
normalized_spans.append(normalized_span)
token_bags.append(set(normalized_span.split()))
return normalized_spans, token_bags
def _align_bags(predicted: List[Set[str]], gold: List[Set[str]]) -> List[float]:
"""
Takes gold and predicted answer sets and first finds the optimal 1-1 alignment
between them and gets maximum metric values over all the answers.
"""
scores = np.zeros([len(gold), len(predicted)])
for gold_index, gold_item in enumerate(gold):
for pred_index, pred_item in enumerate(predicted):
if _match_numbers_if_present(gold_item, pred_item):
scores[gold_index, pred_index] = _compute_f1(pred_item, gold_item)
row_ind, col_ind = linear_sum_assignment(-scores)
max_scores = np.zeros([max(len(gold), len(predicted))])
for row, column in zip(row_ind, col_ind):
max_scores[row] = max(max_scores[row], scores[row, column])
return max_scores
def _compute_f1(predicted_bag: Set[str], gold_bag: Set[str]) -> float:
intersection = len(gold_bag.intersection(predicted_bag))
if not predicted_bag:
precision = 1.0
else:
precision = intersection / float(len(predicted_bag))
if not gold_bag:
recall = 1.0
else:
recall = intersection / float(len(gold_bag))
f1 = (
(2 * precision * recall) / (precision + recall)
if not (precision == 0.0 and recall == 0.0)
else 0.0
)
return f1
def _match_numbers_if_present(gold_bag: Set[str], predicted_bag: Set[str]) -> bool:
gold_numbers = set()
predicted_numbers = set()
for word in gold_bag:
if _is_number(word):
gold_numbers.add(word)
for word in predicted_bag:
if _is_number(word):
predicted_numbers.add(word)
if (not gold_numbers) or gold_numbers.intersection(predicted_numbers):
return True
return False
def get_metrics(
predicted: Union[str, List[str], Tuple[str, ...]],
gold: Union[str, List[str], Tuple[str, ...]],
) -> Tuple[float, float]:
"""
Takes a predicted answer and a gold answer (that are both either a string or a list of
strings), and returns exact match and the DROP F1 metric for the prediction. If you are
writing a script for evaluating objects in memory (say, the output of predictions during
validation, or while training), this is the function you want to call, after using
:func:`answer_json_to_strings` when reading the gold answer from the released data file.
"""
predicted_bags = _answer_to_bags(predicted)
gold_bags = _answer_to_bags(gold)
if set(predicted_bags[0]) == set(gold_bags[0]) and len(predicted_bags[0]) == len(
gold_bags[0]
):
exact_match = 1.0
else:
exact_match = 0.0
f1_per_bag = _align_bags(predicted_bags[1], gold_bags[1])
f1 = np.mean(f1_per_bag)
f1 = round(f1, 2)
return exact_match, f1
def evaluate_strings(prediction, gold):
if type(prediction) != list and type(prediction) != str:
prediction = str(prediction)
if type(gold) != list and type(gold) != str:
gold = str(gold)
try:
predicted_bags = _answer_to_bags(prediction)
gold_bags = _answer_to_bags(gold)
f1_per_bag = _align_bags(predicted_bags[1], gold_bags[1])
f1 = np.mean(f1_per_bag)
except Exception:
f1 = 0.0
return f1 |