-
I am building the Kubeflow Pipeline at that time i am using the eval_config with threshold and then i am creating the pipeline with kubeflowdagrunner and TFX Kubeflow image and running the pipeline i…
-
The component test infra replaces the image tag with the new ones from the cloudbuild. Since now we use the yaml reference instead of the image tag, the component tests need to be fixed to accommodate…
-
**What happened:**
Testing the pipelines notebook examples under Kubeflow 0.7 using Workload Identity on Google Kubernetes Engine.
The [container_build](https://github.com/kubeflow/pipelines/bl…
-
Is there a way to define kubeflow pipelines for training models in other languages (not python)?
-
### What steps did you take:
[A clear and concise description of what the bug is.]
I tried
```
df = pd.read_csv(prediction_result_path)
hello_world = """
Hell…
-
### Background
Currently, kfp manages visualization through a collections of [`viewer` components](https://github.com/kubeflow/pipelines/tree/master/frontend/src/components/viewers).
Ignoring vie…
-
Looking at the current TFX components, it seems like they only run in local mode. I think it would be useful to consider adding support for running on top of Beam w/Dataflow (and later Flink & Spark).
-
Is there a reason for the Python SDK (`kfp`) be in the `pipelines` repository?
Wouldn't it be better to have it's own repository?
-
`BeamDagRunner` mentions the following,
```python
"""
beam_orchestrator_args: beam args for the beam orchestrator. Note that
this is different from the beam_pipeline_args within
…
-
GCP is not accessible in my region, so I have to download the data into the pv and run the taxi sample in local mode [taxi_pipeline_kubeflow_local.py](https://github.com/tensorflow/tfx/blob/master/tfx…