-
The secret to accessing Minio for models is created only if the deployment has MLflow added.
There is a valid use case for deploying a model stored in Minio without MLflow (ie sklearn model hosting…
-
## Describe the bug
Running an experiment with a mirror between pipelines and inspecting each pipeline after sending a single request reveals that the request IDs between the default route and the mi…
-
We could explore extending the current interface to introduce the concept of a `Server` which could abstract the logic related to conda environments, as well as potentially docker images and the workf…
-
I'm trying to run this project and this error keeps happening while trying to execute next command `docker run -p 5001:5000 -it seldon-app`
I haven't changed anything in Dockerfile (tried, but did …
-
How would extensions be handled for existing endpoints such as [model metadata](https://github.com/kubeflow/kfserving/blob/master/docs/predict-api/v2/required_api.md#model-metadata)
An example cou…
-
Install and remove Charmed Kubeflow - multiple CRDs are left:
- istio
- training-job
- seldon
Reproduce:
1. Deploy kubeflow on new cluster
2. destroy model
3. `kubectl get crd`
-
## Describe the bug
From [CVE-2024-1135](https://github.com/advisories/GHSA-w3h3-4rj7-4ph4):
`Gunicorn fails to properly validate Transfer-Encoding headers, leading to HTTP Request Smuggling (HRS) v…
-
Create multi-language wrappers that can run Java, C++ and R models. For this, we can leverage the existing research in Seldon Core which leverages tools like JNI and PyBind to bridge from Python to ot…
-
/kind bug
I am trying to run the ames model prediction using fairing as per the readme at
https://github.com/kubeflow/fairing/tree/master/examples/prediction on cloud and i see pod the follow…
-
## Describe the bug
We want to use the `serviceStatus` field of `status` of a `sdep` resource to get the svc info of all predictors. But sometimes it is missing in the sdep resource. Could y…