Unpin KFP SDK for Kubeflow 1.11.0 and clean up residual dex-based examples

mlflow/mlflow-kfp-example.ipynb

@@ -51,7 +51,8 @@
    },
    "outputs": [],
    "source": [
-    "!pip install -q kfp[all]==2.9.0"
+    "# Following dependency is available by default in prokube notebooks:\n",
+    "# !pip install -q kfp[all]==2.15.0"
    ]
   },
   {

pipelines/lightweight-components/mobile-price-classifications.ipynb

@@ -18,7 +18,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "%pip -q install kagglehub"
+    "%pip -q install kagglehub\n",
+    "# Following dependencies are available by default in prokube notebooks:\n",
+    "# %pip install 'kfp[all]==2.15.0' pyarrow s3fs "
    ]
   },
   {

pipelines/minimal-container-components/README.md

@@ -3,32 +3,4 @@ This example shows a minimal example of how to use
 [container components](https://www.kubeflow.org/docs/components/pipelines/v2/components/container-components/)
 in pipelines.  
 
-The pipeline itself is in `pipeline.py` and you can use `submit-cluster.py` or `submit-remote.py` to submit from
-the cluster (e.g. form a KF-Notebook) or remotely, respectively.
-
-For remote submission, you'll need to follow the preparation steps outlined below.
-
-## Preparing for Remote KFP Connection
-### Deployments that expose pipelines endpoint
-To submit a pipeline remotely you need a deployment that uses dex as an identity provider and exposes the 
-pipelines endopint.
-
-```sh
-export KUBEFLOW_ENDPOINT="http://my-endpoint"
-export KUBEFLOW_USERNAME="example@user.com"
-export KUBEFLOW_PASSWORD="12341234"
- ```
-
-### Deployments that don't expose the endpoint
-In this case we have to resort to port-forwarding
-1. **Port Forwarding**: Set up port forwarding to the cluster by executing the following command:
-   ```sh
-   kubectl port-forward --namespace istio-system svc/istio-ingressgateway 8080:80
-   ```
-   Make sure it remains running in a separate terminal.
-2. **Setting environment variables**: Define these environment variables in your shell. Make sure to replace the username and password with your own credentials:
-    ```sh
-    export KUBEFLOW_ENDPOINT="http://localhost:8080"
-    export KUBEFLOW_USERNAME="example@user.com"
-    export KUBEFLOW_PASSWORD="12341234"
-    ```
+The pipeline itself is in `pipeline.py`. Use `submit-cluster.py` to submit from within the cluster (e.g. from a Kubeflow Notebook).