add cloud-platform scope in the test to reclaim the ai platform sampl…

…e models (#2355)

* add cloud-platform scope
* fix bug in the client wait_for_run_completion

samples/contrib/parameterized_tfx_oss/README.md

@@ -16,6 +16,7 @@ Finally, run `python setup.py install` from `tfx/tfx`. After that, running
 `chicago_taxi_pipeline_simple.py` compiles the TFX pipeline into KFP pipeline package.
 This pipeline requires google storage permission to run. 
 
+
 ## Caveats
 
 This sample uses pipeline parameters in a TFX pipeline, which is not yet fully supported. 

samples/core/ai_platform/ai_platform.ipynb

@@ -280,7 +280,23 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "kfp.Client().create_run_from_pipeline_func(pipeline, arguments={}, experiment_name=experiment_name)"
+    "pipeline = kfp.Client().create_run_from_pipeline_func(pipeline, arguments={}, experiment_name=experiment_name)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Wait for the pipeline to finish"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "pipeline.wait_for_run_completion(timeout=1800)"
    ]
   },
   {

samples/core/xgboost_training_cm/README.md

@@ -12,6 +12,7 @@ or not.
 
 Preprocessing uses Google Cloud DataProc. Therefore, you must enable the [DataProc API](https://cloud.google.com/endpoints/docs/openapi/enable-api) for the given GCP project.
 
+
 ## Compile
 
 Follow the guide to [building a pipeline](https://www.kubeflow.org/docs/guides/pipelines/build-pipeline/) to install the Kubeflow Pipelines SDK and compile the sample Python into a workflow specification. The specification takes the form of a YAML file compressed into a `.zip` file. 

sdk/python/kfp/_client.py

@@ -358,7 +358,7 @@ def __init__(self, client, run_info):
 
       def wait_for_run_completion(self, timeout=None):
         timeout = timeout or datetime.datetime.max - datetime.datetime.min
-        return self._client.wait_for_run_completion(timeout)
+        return self._client.wait_for_run_completion(self.run_id, timeout)
 
       def __str__(self):
         return '<RunPipelineResult(run_id={})>'.format(self.run_id)

test/deploy-cluster.sh

@@ -48,7 +48,7 @@ else
   SHOULD_CLEANUP_CLUSTER=true
   # "storage-rw" is needed to allow VMs to push to gcr.io
   # reference: https://cloud.google.com/compute/docs/access/service-accounts#accesscopesiam
-  SCOPE_ARG="--scopes=storage-rw"
+  SCOPE_ARG="--scopes=storage-rw,cloud-platform"
   # Machine type and cluster size is the same as kubeflow deployment to
   # easily compare performance. We can reduce usage later.
   NODE_POOL_CONFIG_ARG="--num-nodes=2 --machine-type=n1-standard-8 \