Issue: Cannot see scores in LangSmith dashboard

[jasonkang14]

Issue you'd like to raise.

I cannot see the scores of my evaluations in LangSmith experiments dashboard

Below is the code,

# Grade prompt
from langsmith import EvaluationResult
grade_prompt_answer_helpfulness = prompt = hub.pull("langchain-ai/rag-answer-helpfulness")

def answer_helpfulness_evaluator(run, example) -> dict:
    """
    A simple evaluator for RAG answer helpfulness
    """

    # Get question, ground truth answer, RAG chain answer
    input_question = example.inputs.get("input_question") or ""
    prediction = run.outputs.get("output") or ""

    # LLM grader
    llm = ChatOpenAI(model="gpt-4o", temperature=0)

    # Structured prompt
    answer_grader = grade_prompt_answer_helpfulness | llm

    # Run evaluator
    score = answer_grader.invoke({"question": input_question,
                                  "student_answer": prediction})
    score = score["Score"]
    print(f'score in answer_helpfulness_evaluator: {score}')
    return EvaluationResult(key="answer_helpfulness_score", score=score)

from uuid import uuid4
from langsmith.evaluation import evaluate
from llm import get_ai_response

experiment_results = evaluate(
    lambda inputs: get_ai_response(inputs["input_question"]),
    data=dataset_name, 
    max_concurrency=2, 
    evaluators=[answer_evaluator, answer_helpfulness_evaluator], 
    experiment_prefix="test-run",
    metadata={"version": str(uuid4()), "chunk_count": "2"}, 
)

Suggestion:

Am I missing a configuration?

[hinthornw]

Hm thislooks correct at first glance - are there any error logs in your terminal?

[jasonkang14]

i dont see any error logs in the terminal.