Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

This commit implements the F-beta score metric #1543

Merged
merged 8 commits into from
Oct 25, 2024
Merged

This commit implements the F-beta score metric #1543

Changes from 1 commit
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
e1f0ca8
This commit implements the F-beta score metric
Yuri-Albuquerque Oct 20, 2024
1c339e9
also implements the F-beta score for
Yuri-Albuquerque Oct 21, 2024
ed3b3f3
returning to the original recall and precision
Yuri-Albuquerque Oct 22, 2024
92166c0
remove ALL_METRICS
shahules786 Oct 25, 2024
69cfa99
add fbeta score
shahules786 Oct 25, 2024
a0f6917
replace by fbeta score
shahules786 Oct 25, 2024
c432b19
removed unused imports
shahules786 Oct 25, 2024
8d5050c
Merge branch 'main' into feature/f-beta-score
shahules786 Oct 25, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 6 additions & 1 deletion src/ragas/metrics/_answer_correctness.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -167,6 +167,7 @@ class AnswerCorrectness(MetricWithLLM, MetricWithEmbeddings, SingleTurnMetric):
default_factory=LongFormAnswerPrompt
)
weights: list[float] = field(default_factory=lambda: [0.75, 0.25])
beta: float = field(default_factory= lambda: 1.0)
answer_similarity: t.Optional[AnswerSimilarity] = None
sentence_segmenter: t.Optional[HasSegmentMethod] = None
max_retries: int = 1
Expand All @@ -185,6 +186,9 @@ def __post_init__(self: t.Self):
language = self.long_form_answer_prompt.language
self.sentence_segmenter = get_segmenter(language=language, clean=False)

if type(self.beta) is not float:
raise ValueError("Beta must be a float. A beta > 1 gives more weight to recall, while beta < 1 favors precision.")

def init(self, run_config: RunConfig):
super().init(run_config)
if self.answer_similarity is None and self.weights[1] != 0:
Expand All @@ -198,7 +202,8 @@ def _compute_statement_presence(
tp = len(prediction.TP)
fp = len(prediction.FP)
fn = len(prediction.FN)
score = tp / (tp + 0.5 * (fp + fn)) if tp > 0 else 0
beta = self.beta
score = ((1 + beta * beta) * tp) / ((1 + beta * beta) * tp + fp + (beta * beta) * fn) if tp > 0 else 0
return score

async def _create_simplified_statements(
Expand Down
Loading