Add AnswerExactMatchEvaluator

[silvanocerza]

Related Issues

• fixes Implement function to calculate Exact Match metric #6067

Proposed Changes:

Add AnswerExactMatchEvaluator. This Component calculates the Exact Match metrics given a list of questions, a list of expected answers for each question and the list of predicted answers for each question.

How did you test it?

I added unit tests.

Notes for the reviewer

N/A

Checklist

• I have read the contributors guidelines and the code of conduct
• I have updated the related issue with new insights and changes
• I added unit tests and updated the docstrings
• I've used one of the conventional commit types for my PR title: fix:, feat:, build:, chore:, ci:, docs:, style:, refactor:, perf:, test:.
• I documented my code
• I ran pre-commit hooks and fixed any issue

[julian-risch]

I believe we can leave out from_dict and to_dict implementations as they have the same effect as the default implementation. Otherwise looks good to me.

[julian-risch]

We can leave out the to_dict and from_dict implementation here as it is just using the default right?
For example, when component_to_dict is used it will automatically fall back to default_to_dict:

haystack/haystack/core/serialization.py

Line 10 in f69c3e5

def component_to_dict(obj: Any) -> Dict[str, Any]:

[silvanocerza]

Right, I always forget that. Will remove them right away.

[julian-risch]

LGTM! 👍
We should add more test cases later. Will also make sense for consistency with test cases for other metrics. For example, we could test more than one prediction per query. Something like:

evaluator.run(
    questions=["What is the capital of Germany?", "What is the capital of France?"],
    ground_truth_answers=[["Berlin"], ["London"]],
    predicted_answers=[["Berlin", "wrong_second_answer_candidate"], ["wrong_first_answer_candidate", "London"]],
)

should result in result["result"] == 1.0

[coveralls]

Pull Request Test Coverage Report for Build 8345123752

Details

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.03%) to 89.238%

---

Totals
Change from base Build 8339327256:	0.03%
Covered Lines:	5390
Relevant Lines:	6040

---

💛 - Coveralls