where calculate_em_str_multi define?

[LiemLin]

PaddleNLP/pipelines/pipelines/pipelines/base.py

Line 589 in 31a8dc5

lambda row: calculate_em_str_multi(

[w5688414]

I am not sure why you meet many problems for the missing code, could you tell me more details why you met such problems.

def calculate_em_str_multi(gold_labels, prediction):
    for gold_label in gold_labels:
        result = int(normalize_answer(gold_label) == normalize_answer(prediction))
        if result == 1.0:
            return 1.0
    return 0.0

def normalize_answer(s: str):
    """
    Lower text and remove punctuation, articles and extra whitespace.
    """

    def remove_articles(text):
        regex = re.compile(r"\b(a|an|the)\b", re.UNICODE)
        return re.sub(regex, " ", text)

    def white_space_fix(text):
        return " ".join(text.split())

    def remove_punc(text):
        exclude = set(string.punctuation)
        return "".join(ch for ch in text if ch not in exclude)

    def lower(text):
        return text.lower()

    return white_space_fix(remove_articles(remove_punc(lower(s))))

[LiemLin]

git clone https://github.com/PaddlePaddle/PaddleNLP.git

When I opened the project, I found that many methods were missing definitions

[w5688414]

git clone https://github.com/PaddlePaddle/PaddleNLP.git

When I opened the project, I found that many methods were missing definitions

Which IDE did you use?

[LiemLin]

Pycharm at Mac OS

def calculate_em_str_multi(gold_labels, prediction):
for gold_label in gold_labels:
result = int(normalize_answer(gold_label) == normalize_answer(prediction))
if result == 1.0:
return 1.0
return 0.0

Do I put this function define in pipelines/base.py ?

[LiemLin]

and this function calculate_f1_str_multi no define yet

[w5688414]

or you can remove this function like this pr #3330

import collections
import re
import string

def get_tokens(s: str):
    if not s:
        return []
    return normalize_answer(s).split()


def compute_f1(a_gold: str, a_pred: str):
    gold_toks = get_tokens(a_gold)
    pred_toks = get_tokens(a_pred)
    common: collections.Counter = collections.Counter(gold_toks) & collections.Counter(pred_toks)
    num_same = sum(common.values())
    if len(gold_toks) == 0 or len(pred_toks) == 0:
        # If either is no-answer, then F1 is 1 if they agree, 0 otherwise
        return int(gold_toks == pred_toks)
    if num_same == 0:
        return 0
    precision = 1.0 * num_same / len(pred_toks)
    recall = 1.0 * num_same / len(gold_toks)
    f1 = (2 * precision * recall) / (precision + recall)
    return f1

def calculate_f1_str_multi(gold_labels, prediction):
    results = []
    for gold_label in gold_labels:
        result = compute_f1(gold_label, prediction)
        results.append(result)
    if len(results) > 0:
        return max(results)
    else:
        return 0.0

[LiemLin]

ok thanks