Christian Ritter
Christian Ritter
#### Describe the bug Kubeflow Pipelines from other user namespace are visible under Pipelines in the Kubeflow UI. This includes the yaml code. I would argue it would be better...
In this epic we continue (previous steps at #699) re-enabling Seldon Core on the production cluster and to allow for isolated user-namespace deployments. Sub-issues can be linked against this main...
#### Describe the bug Using VS Code in VNC notebook servers we found that it often crashes due to unknown reasons, not providing any error message. Crashes frequently, on multiple...
The integration of data validation step via Kubeflow Pipelines as part of a MLOps system is of great benefit. The open source tool GreatExpectations as a data validation tool (incl....
Hello, [chassis](https://github.com/modzy/chassis) is an option to build docker images based on machine learning models using Kaniko. It could be an option to build docker images on AAW namespaces.
Seldon Core supports autoscaling [doc](https://docs.seldon.io/projects/seldon-core/en/latest/examples/autoscaling_example.html) and the functionality would need to be tested, benchmarke. e.g. similar in approach to https://arxiv.org/abs/2202.11243, albeit we don't have smart batching.
Hi @ianepreston, Functions such as `vectors_to_df` accept date-related parameters strings in the format YYYY-MM-DD which are more convenient for users to pass to functions than `datetime.date` objects. One could add...
Currently, Seldon deployments are not accessible due to network restrictions and require a special network setup, as identified and used in #798. The network settings need to be adapted for...