search

SeldonIO / seldon-core

An MLOps framework to package, deploy, monitor and manage thousands of production machine learning models
https://www.seldon.io/tech/products/core/
Other
4.32k stars 826 forks source link

Relax PyYAML constraints in MLflow server #5426

Open jtisbell4 opened 5 months ago

jtisbell4 commented 5 months ago

I am getting a failure in my deployment which I think is related to #5091 (I am using seldon-core v1.18)

I believe it's tied to the following: https://github.com/SeldonIO/seldon-core/blob/6a45b2b0d803fdce8761d40639566db5272db065/servers/mlflowserver/mlflowserver/requirements.txt#L2

Can we please get these constraints relaxed to allow for PyYAML>=6?

PennyXQZ commented 2 months ago

I meet the same mismatch warning as follow:

2024/06/05 14:33:21 WARNING mlflow.utils.requirements_utils: Detected one or more mismatches between the model's dependencies and the current Python environment:
 - pyyaml (current: 5.4.1, required: pyyaml==6.0.1)
To fix the mismatches, call `mlflow.pyfunc.get_model_dependencies(model_uri)` to fetch the model's environment and install dependencies using the resulting environment file.
2024/06/05 14:33:21 WARNING mlflow.utils.requirements_utils: Detected one or more mismatches between the model's dependencies and the current Python environment:
 - pyyaml (current: 5.4.1, required: pyyaml==6.0.1)
To fix the mismatches, call `mlflow.pyfunc.get_model_dependencies(model_uri)` to fetch the model's environment and install dependencies using the resulting environment file.

'pyyaml==6.0.1' was collected and built during conda env setting up at first, but degraded to 'PyYAML-5.4.1' as limitation by 'requirements.txt'.