Scripts to create a Dbizi dataset and evaluate an assistant#227
Merged
juananpe merged 4 commits intoLamb-Project:langsmithfrom Jan 29, 2026
Merged
Scripts to create a Dbizi dataset and evaluate an assistant#227juananpe merged 4 commits intoLamb-Project:langsmithfrom
juananpe merged 4 commits intoLamb-Project:langsmithfrom
Conversation
Contributor
Author
|
Fixed issues in evaluate_assistant.py and added dbizi_dataset_eus.py (same as dizi_dataset.py but in basque). |
Contributor
Author
|
Now these variables are loaded from lamb-kb-server-stable/backend/.env instead of having them hardcoded: |
Contributor
Author
|
Now EVALUATOR_MODEL is chosen via env var. |
juananpe
approved these changes
Jan 29, 2026
| load_dotenv(project_root / "backend" / ".env") | ||
|
|
||
| # Configuration | ||
| API_BASE_URL = "http://localhost:9099" |
Collaborator
There was a problem hiding this comment.
Please, use env variables here to avoid hardcoding them
| try: | ||
| # Call the LLM judge (gpt-4.1) | ||
| response = openai_client.chat.completions.create( | ||
| model="gpt-4.1", |
Collaborator
There was a problem hiding this comment.
should be possible to define the model via env vars
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Added 2 scripts in scripts/langsmith:
-dbizi_dataset.py (creates a dataset in LangSmith with 15 questions about dbizi)
-evaluate_assistant.py (an llm-as-a-judge evaluates a given lamb assistant with the given dataset in LangSmith)
To try them, first configure these LangSmith env variables in backend/.env:
-LANGCHAIN_TRACING_V2=true
-LANGCHAIN_API_KEY=YOUR_LANGCHAIN_API_KEY
-LANGCHAIN_PROJECT=lamb-assistants
-LANGCHAIN_ENDPOINT=https://api.smith.langchain.com
And these 2 variables in evaluate_assistant.py:
-JWT_TOKEN = "your_jwt_token_here"
-ASSISTANT_ID = 1 (this should be the id of the dbizi assistant in lamb)
Then, run the scripts:
-first, dbizi_dataset.py
-then, evaluate_assistant.py