traceloop / traceloop/openllmetry
🚀 Feature: Support for mlflow
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 7.4k
- Forks
- 1.1k
- Avg merge
- 8d 14h
- Merged PRs (30d)
- 2
Description
Which component is this feature for?
OpenAI Instrumentation
🔖 Feature description
In mlflow.evaluate and mlflow.predict with OpenAI models, I can't see traces.
If I save the model and load it to use for prediction, I can't see traces as well. It would be better if you had something like MlflowInstrument or something similar.
🎤 Why is this feature needed ?
I always use MLflow with TraceLoop, but it always turns out that I cant see traces.
✌️ How do you aim to achieve this?
Support MLflow features such as predict or evaluate, especially with OpenAI.
🔄️ Additional Information
Prior bug: https://github.com/traceloop/openllmetry/issues/1750
👀 Have you spent some time to check if this feature request has been raised before?
- I checked and didn't find similar issue
Are you willing to submit PR?
None
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reviewing the OpenAI Instrumentation component and the MLflow workflows named in the issue: mlflow.evaluate, mlflow.predict, and prediction after saving and loading a model. Compare these cases with the prior bug in issue 1750; done means traces are visible for MLflow prediction and evaluation with OpenAI models.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning, observability
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100