"docling>=2.36.1",
    "giskard[llm]>=2.17.0",
    "ipykernel>=6.29.5",
    "jupyter>=1.1.1",
    "llama-index>=0.12.41",
    "llama-index-embeddings-azure-openai>=0.3.8",
    "llama-index-embeddings-ollama>=0.6.0",
    "llama-index-llms-azure-openai>=0.3.4",
    "llama-index-node-parser-docling>=0.3.2",
    "llama-index-readers-docling>=0.3.2",
    "python-dotenv>=1.1.0",
    "ragas>=0.2.15",

def answer_fn(
    question: str,
    history: list[dict] = None,
) -> AgentAnswer:
    if history:
        answer = chat_engine.chat(
            question,
            chat_history=[
                ChatMessage(**chat_history_dict) for chat_history_dict in history
            ],
        )
    else:
        answer = chat_engine.chat(question, chat_history=[])
    return AgentAnswer(
        message = str(answer),
        documents=[source.content for source in answer.sources]
    )


report = evaluate(
    answer_fn, testset=testset, 
    knowledge_base=knowledge_base,
    metrics=[ragas_context_recall, ragas_context_precision],
)


I expected this to generate a comprehensive report evaluating the RAG application

...
File ~/bug-free-octo-pancake/.venv/lib/python3.12/site-packages/giskard/rag/metrics/ragas_metrics.py:98, in RagasMetric.__call__(self, question_sample, answer)
     96     self.ragas_llm = RagasLLMWrapper(llm_client, self.context_window_length)
     97 if self.ragas_embeddings is None:
---> 98     self.ragas_embeddings = RagasEmbeddingsWrapper(embedding_model)
    100 run_config = RunConfig()
    102 if hasattr(self.metric, "llm"):

TypeError: Can't instantiate abstract class RagasEmbeddingsWrapper without an implementation for abstract methods 'aembed_documents', 'aembed_query'


This is because the `RagasEmbeddingsWrapper` class in giskard > rag > metrics > ragas_metrics.py does not implement the abstract methods from the `BaseRagasEmbeddings` class:

class RagasEmbeddingsWrapper(BaseRagasEmbeddings):
    def __init__(self, embedding_model):
        self.embedding_model = embedding_model

    def embed_query(self, text: str) -> Sequence[float]:
        return self.embedding_model.embed([text])[0]

    def embed_documents(self, texts: Sequence[str]) -> Sequence[Sequence[float]]:
        return self.embedding_model.embed(texts)