GIS KFP example: kfp-related syntax changes; use gcs client libs instead of gsutil (kubeflow#749)

* some mods to accommodate (perhaps temporary) changes in how the kfp sdk works

* Use gcs client libs rather than gsutil for a gcs copy; required due to changes in node service account permissions.

* more mods to address kfp syntax changes

Author: amygdala

github_issue_summarization/pipelines/components/t2t/t2t-proc/datagen.py

@@ -39,6 +39,16 @@ def copy_local_directory_to_gcs(project, local_path, bucket_name, gcs_path):
     blob = bucket.blob(remote_path)
     blob.upload_from_filename(local_file)
 
+def download_blob(bucket_name, source_blob_name, destination_file_name):
+    """Downloads a blob from the bucket."""
+    storage_client = storage.Client()
+    bucket = storage_client.bucket(bucket_name)
+    blob = bucket.blob(source_blob_name)
+    blob.download_to_filename(destination_file_name)
+    print("Blob {} downloaded to {}.".format(
+            source_blob_name, destination_file_name)
+    )
+
 def main():
   parser = argparse.ArgumentParser(description='ML Trainer')
   parser.add_argument(
@@ -57,13 +67,8 @@ def main():
   local_data_dir = '/ml/t2t_gh_data'
   local_source_data_file = '/ml/gh_data/github_issues.csv'
 
-  data_copy_command1 = ['gsutil', 'cp',
-      'gs://aju-dev-demos-codelabs/kubecon/gh_data/github_issues.csv',
-      local_source_data_file
-      ]
-  print(data_copy_command1)
-  result = subprocess.call(data_copy_command1)
-  print(result)
+  download_blob('aju-dev-demos-codelabs', 'kubecon/gh_data/github_issues.csv',
+      local_source_data_file)
 
   datagen_command = ['t2t-datagen', '--data_dir', local_data_dir, '--t2t_usr_dir',
       '/ml/ghsumm/trainer',

github_issue_summarization/pipelines/example_pipelines/gh_summ.py

@@ -16,7 +16,7 @@
 import kfp.dsl as dsl
 import kfp.gcp as gcp
 import kfp.components as comp
-from kfp.dsl.types import GCSPath, String
+# from kfp.dsl.types import GCSPath, String
 
 
 COPY_ACTION = 'copy_data'
@@ -43,12 +43,12 @@
 )
 def gh_summ(  #pylint: disable=unused-argument
   train_steps: 'Integer' = 2019300,
-  project: String = 'YOUR_PROJECT_HERE',
-  github_token: String = 'YOUR_GITHUB_TOKEN_HERE',
-  working_dir: GCSPath = 'gs://YOUR_GCS_DIR_HERE',
-  checkpoint_dir: GCSPath = 'gs://aju-dev-demos-codelabs/kubecon/model_output_tbase.bak2019000/',
-  deploy_webapp: String = 'true',
-  data_dir: GCSPath = 'gs://aju-dev-demos-codelabs/kubecon/t2t_data_gh_all/'
+  project: str = 'YOUR_PROJECT_HERE',
+  github_token: str = 'YOUR_GITHUB_TOKEN_HERE',
+  working_dir: 'GCSPath' = 'gs://YOUR_GCS_DIR_HERE',
+  checkpoint_dir: 'GCSPath' = 'gs://aju-dev-demos-codelabs/kubecon/model_output_tbase.bak2019000/',
+  deploy_webapp: str = 'true',
+  data_dir: 'GCSPath' = 'gs://aju-dev-demos-codelabs/kubecon/t2t_data_gh_all/'
   ):
 
 

github_issue_summarization/pipelines/example_pipelines/gh_summ.py.tar.gz

@@ -15,15 +15,15 @@
 
 import kfp.dsl as dsl
 import kfp.gcp as gcp
-from kfp.dsl.types import String
+# from kfp.dsl.types import String
 
 
 @dsl.pipeline(
   name='Github issue summarization',
   description='Demonstrate Tensor2Tensor-based training and TF-Serving'
 )
 def gh_summ_serveonly(
-  github_token: String = 'YOUR_GITHUB_TOKEN_HERE',
+  github_token: str = 'YOUR_GITHUB_TOKEN_HERE',
   ):
 
 

github_issue_summarization/pipelines/example_pipelines/gh_summ_serve.py

@@ -173,12 +173,12 @@
     ")\n",
     "def gh_summ(  #pylint: disable=unused-argument\n",
     "  train_steps: 'Integer' = 2019300,\n",
-    "  project: String = 'YOUR_PROJECT_HERE',\n",
-    "  github_token: String = 'YOUR_GITHUB_TOKEN_HERE',\n",
-    "  working_dir: GCSPath = 'YOUR_GCS_DIR_HERE',\n",
-    "  checkpoint_dir: GCSPath = 'gs://aju-dev-demos-codelabs/kubecon/model_output_tbase.bak2019000/',\n",
-    "  deploy_webapp: String = 'true',\n",
-    "  data_dir: GCSPath = 'gs://aju-dev-demos-codelabs/kubecon/t2t_data_gh_all/'\n",
+    "  project: str = 'YOUR_PROJECT_HERE',\n",
+    "  github_token: str = 'YOUR_GITHUB_TOKEN_HERE',\n",
+    "  working_dir: 'GCSPath' = 'YOUR_GCS_DIR_HERE',\n",
+    "  checkpoint_dir: 'GCSPath' = 'gs://aju-dev-demos-codelabs/kubecon/model_output_tbase.bak2019000/',\n",
+    "  deploy_webapp: str = 'true',\n",
+    "  data_dir: 'GCSPath' = 'gs://aju-dev-demos-codelabs/kubecon/t2t_data_gh_all/'\n",
     "  ):\n",
     "\n",
     "\n",
@@ -266,8 +266,8 @@
    "source": [
     "# You'd uncomment this call to actually run the pipeline. \n",
     "# run = client.run_pipeline(exp.id, 'ghsumm', 'ghsumm.tar.gz',\n",
-    "#                           params={'working-dir': WORKING_DIR,\n",
-    "#                                   'github-token': GITHUB_TOKEN,\n",
+    "#                           params={'working_dir': WORKING_DIR,\n",
+    "#                                   'github_token': GITHUB_TOKEN,\n",
     "#                                   'project': PROJECT_NAME})"
    ]
   },
@@ -329,12 +329,12 @@
     ")\n",
     "def gh_summ2(\n",
     "  train_steps: 'Integer' = 2019300,\n",
-    "  project: String = 'YOUR_PROJECT_HERE',\n",
-    "  github_token: String = 'YOUR_GITHUB_TOKEN_HERE',\n",
-    "  working_dir: GCSPath = 'YOUR_GCS_DIR_HERE',\n",
-    "  checkpoint_dir: GCSPath = 'gs://aju-dev-demos-codelabs/kubecon/model_output_tbase.bak2019000/',\n",
-    "  deploy_webapp: String = 'true',\n",
-    "  data_dir: GCSPath = 'gs://aju-dev-demos-codelabs/kubecon/t2t_data_gh_all/'\n",
+    "  project: str = 'YOUR_PROJECT_HERE',\n",
+    "  github_token: str = 'YOUR_GITHUB_TOKEN_HERE',\n",
+    "  working_dir: 'GCSPath' = 'YOUR_GCS_DIR_HERE',\n",
+    "  checkpoint_dir: 'GCSPath' = 'gs://aju-dev-demos-codelabs/kubecon/model_output_tbase.bak2019000/',\n",
+    "  deploy_webapp: str = 'true',\n",
+    "  data_dir: 'GCSPath' = 'gs://aju-dev-demos-codelabs/kubecon/t2t_data_gh_all/'\n",
     "  ):\n",
     "\n",
     "  # The new pre-processing op.\n",
@@ -418,9 +418,9 @@
    "outputs": [],
    "source": [
     "run = client.run_pipeline(exp.id, 'ghsumm2', 'ghsumm2.tar.gz',\n",
-    "                          params={'working-dir': WORKING_DIR,\n",
-    "                                  'github-token': GITHUB_TOKEN,\n",
-    "                                  'deploy-webapp': DEPLOY_WEBAPP,\n",
+    "                          params={'working_dir': WORKING_DIR,\n",
+    "                                  'github_token': GITHUB_TOKEN,\n",
+    "                                  'deploy_webapp': DEPLOY_WEBAPP,\n",
     "                                  'project': PROJECT_NAME})"
    ]
   },