Heuristic Metrics

.. currentmodule:: opik.evaluation.metrics

This section lists the text-similarity, readability, safety, and sentiment metrics that can be composed into larger test suites. Several metrics rely on optional dependencies (for example, bert-score, sacrebleu, fasttext, nltk); install the relevant packages before scoring.

Sentence & Token Overlap

.. autoclass:: SentenceBLEU :special-members: init :members: score

.. autoclass:: CorpusBLEU :special-members: init :members: score

.. autoclass:: GLEU :special-members: init :members: score

.. autoclass:: ROUGE :special-members: init :members: score

.. autoclass:: ChrF :special-members: init :members: score

.. autoclass:: METEOR :special-members: init :members: score

.. autoclass:: BERTScore :special-members: init :members: score

Distribution Comparisons

.. autoclass:: JSDivergence :special-members: init :members: score

.. autoclass:: JSDistance :special-members: init :members: score

.. autoclass:: KLDivergence :special-members: init :members: score

Rank & Readability

.. autoclass:: SpearmanRanking :special-members: init :members: score

.. autoclass:: Readability :special-members: init :members: score

.. autoclass:: Tone :special-members: init :members: score

Prompt Safety & Sentiment

.. autoclass:: PromptInjection :special-members: init :members: score

.. autoclass:: LanguageAdherenceMetric :special-members: init :members: score

.. autoclass:: Sentiment :special-members: init :members: score

.. autoclass:: VADERSentiment :special-members: init :members: score