Marcin Zabłocki
Marcin Zabłocki
## Description There is a `mlflow-skinny` package which contains all of the tracking features with minimal set of dependencies. Have you considered using it in `kedro-mlflow`? Regardless of the answer...
**Description of your changes:** Add support for Agro's continueOn capability to Kubeflow Pipelines. This allows the pipelines to continue execution even though some of the jobs fail, which is really...
## Description In many of our integrations we deploy MLflow instances behind some authorization methods, is most cases - an OAuth2.0. Secured MLflow instances require HTTP requests to have the...
## Description I have observed an unexpected behavior related to namespace filtering in modular pipelines. ## Context Specifically, when I have two modular pipelines under the same namespace, such as:...
The example linked in `README` https://github.com/kubeflow/xgboost-operator/tree/master/config/samples/xgboost-dist shows that spawning distributed training job requires running `kubectl`. I want to run distributed XGBoost training as a part of bigger Kubeflow pipeline, how...
## Description I have a project where there is a huge number of pipelines generated programatically (in a loop). The process of generating those pipelines takes a lot of time...
This PR creates e2e tests. They will most likely fail, due to many changes introduced in #144 . --- Keep in mind: ~~- [ ] Documentation updates~~ N/A - [x]...
Right now, e2e test don't use `TemplatedConfigLoader` by default and the `${run_id}` interpolation does not work properly in the https://github.com/getindata/kedro-kubeflow/blob/d3cff89ce03e9bcc1b31598a8fa17aadeb2717ab/tests/e2e/catalog.yml#L20 Possible fix: Use TemplatedConfigLoader in e2e tests
The latest update to `kedro-docker` (which we've introduced BTW 😀) replaced `/home/kedro` with `/home/kedro_docker` due to name collision if someone had `__init__.py` in their project directory - which is a...
Right now, there is no support for kedro-mlflow as it handles the mlflow run initialization on it's own via kedro hooks. In Azure ML Pipelines, at runtime, all of the...