Remove BuildId parameter (microsoft#214)

Remove the need to add the BuildId parameter and tagging logic to the train/evaluate scripts. The tags are instead specified when the experiment is submitted. The register script pulls the BuildId from the tag.

Author: jotaylo

.pipelines/diabetes_regression-ci.yml

@@ -50,6 +50,8 @@ stages:
 - stage: 'Trigger_AML_Pipeline'
   displayName: 'Train model'
   condition: and(succeeded(), not(variables['MODEL_BUILD_ID']))
+  variables:
+    BUILD_URI: '$(SYSTEM.COLLECTIONURI)$(SYSTEM.TEAMPROJECT)/_build/results?buildId=$(BUILD.BUILDID)'
   jobs:
   - job: "Get_Pipeline_ID"
     condition: and(succeeded(), eq(coalesce(variables['auto-trigger-training'], 'true'), 'true'))
@@ -85,7 +87,7 @@ stages:
         azureSubscription: '$(WORKSPACE_SVC_CONNECTION)'
         PipelineId: '$(AMLPIPELINE_ID)'
         ExperimentName: '$(EXPERIMENT_NAME)'
-        PipelineParameters: '"ParameterAssignments": {"model_name": "$(MODEL_NAME)"}'
+        PipelineParameters: '"ParameterAssignments": {"model_name": "$(MODEL_NAME)"}, "tags": {"BuildId": "$(Build.BuildId)", "BuildUri": "$(BUILD_URI)"}, "StepTags": {"BuildId": "$(Build.BuildId)", "BuildUri": "$(BUILD_URI)"}'
   - job: "Training_Run_Report"
     dependsOn: "Run_ML_Pipeline"
     condition: always()

diabetes_regression/evaluate/evaluate_model.py

@@ -23,7 +23,6 @@
 ARISING IN ANY WAY OUT OF THE USE OF THE SOFTWARE CODE, EVEN IF ADVISED OF THE
 POSSIBILITY OF SUCH DAMAGE.
 """
-import os
 from azureml.core import Run
 import argparse
 import traceback
@@ -74,11 +73,7 @@
 run_id = 'amlcompute'
 
 parser = argparse.ArgumentParser("evaluate")
-parser.add_argument(
-    "--build_id",
-    type=str,
-    help="The Build ID of the build triggering this pipeline run",
-)
+
 parser.add_argument(
     "--run_id",
     type=str,
@@ -99,19 +94,13 @@
 )
 
 args = parser.parse_args()
-if (args.build_id is not None):
-    build_id = args.build_id
 if (args.run_id is not None):
     run_id = args.run_id
 if (run_id == 'amlcompute'):
     run_id = run.parent.id
 model_name = args.model_name
 metric_eval = "mse"
-run.tag("BuildId", value=build_id)
-builduri_base = os.environ.get("BUILDURI_BASE")
-if (builduri_base is not None):
-    build_uri = builduri_base + build_id
-    run.tag("BuildUri", value=build_uri)
+
 allow_run_cancel = args.allow_run_cancel
 # Parameterize the matrices on which the models should be compared
 # Add golden data set on which all the model performance can be evaluated

diabetes_regression/register/register_model.py

@@ -43,7 +43,6 @@ def main():
         experiment_name = os.environ.get("EXPERIMENT_NAME")
         resource_group = os.environ.get("RESOURCE_GROUP")
         subscription_id = os.environ.get("SUBSCRIPTION_ID")
-        build_id = os.environ.get('BUILD_BUILDID')
         # run_id useful to query previous runs
         run_id = "bd184a18-2ac8-4951-8e78-e290bef3b012"
         aml_workspace = Workspace.get(
@@ -59,11 +58,6 @@ def main():
         run_id = 'amlcompute'
 
     parser = argparse.ArgumentParser("register")
-    parser.add_argument(
-        "--build_id",
-        type=str,
-        help="The Build ID of the build triggering this pipeline run",
-    )
 
     parser.add_argument(
         "--run_id",
@@ -84,8 +78,6 @@ def main():
     )
 
     args = parser.parse_args()
-    if (args.build_id is not None):
-        build_id = args.build_id
     if (args.run_id is not None):
         run_id = args.run_id
     if (run_id == 'amlcompute'):
@@ -98,32 +90,40 @@ def main():
     model_file = os.path.join(model_path, model_name)
     model = joblib.load(model_file)
     model_mse = run.parent.get_metrics()["mse"]
+    parent_tags = run.parent.get_tags()
+    try:
+        build_id = parent_tags["BuildId"]
+    except KeyError:
+        build_id = None
+        print("BuildId tag not found on parent run.")
+        print("Tags present: {parent_tags}")
+    try:
+        build_uri = parent_tags["BuildUri"]
+    except KeyError:
+        build_uri = None
+        print("BuildUri tag not found on parent run.")
+        print("Tags present: {parent_tags}")
 
     if (model is not None):
         if (build_id is None):
-            register_aml_model(model_file, model_name, exp, run_id)
+            register_aml_model(model_file, model_name, model_mse, exp, run_id)
+        elif (build_uri is None):
+            register_aml_model(
+                model_file,
+                model_name,
+                model_mse,
+                exp,
+                run_id,
+                build_id)
         else:
-            run.tag("BuildId", value=build_id)
-            builduri_base = os.environ.get("BUILDURI_BASE")
-            if (builduri_base is not None):
-                build_uri = builduri_base + build_id
-                run.tag("BuildUri", value=build_uri)
-                register_aml_model(
-                    model_file,
-                    model_name,
-                    model_mse,
-                    exp,
-                    run_id,
-                    build_id,
-                    build_uri)
-            else:
-                register_aml_model(
-                    model_file,
-                    model_name,
-                    model_mse,
-                    exp,
-                    run_id,
-                    build_id)
+            register_aml_model(
+                model_file,
+                model_name,
+                model_mse,
+                exp,
+                run_id,
+                build_id,
+                build_uri)
     else:
         print("Model not found. Skipping model registration.")
         sys.exit(0)

diabetes_regression/training/train.py

@@ -50,11 +50,7 @@ def main():
     print("Running train.py")
 
     parser = argparse.ArgumentParser("train")
-    parser.add_argument(
-        "--build_id",
-        type=str,
-        help="The build ID of the build triggering this pipeline run",
-    )
+
     parser.add_argument(
         "--model_name",
         type=str,
@@ -70,12 +66,10 @@ def main():
 
     args = parser.parse_args()
 
-    print("Argument [build_id]: %s" % args.build_id)
     print("Argument [model_name]: %s" % args.model_name)
     print("Argument [step_output]: %s" % args.step_output)
 
     model_name = args.model_name
-    build_id = args.build_id
     step_output_path = args.step_output
 
     print("Getting training parameters")
@@ -119,15 +113,7 @@ def main():
     output_path = os.path.join('outputs', model_name)
     joblib.dump(value=reg, filename=output_path)
 
-    # Add properties to identify this specific training run
-    run.parent.tag("BuildId", value=build_id)
-    run.tag("BuildId", value=build_id)
     run.tag("run_type", value="train")
-    builduri_base = os.environ.get("BUILDURI_BASE")
-    if (builduri_base is not None):
-        build_uri = builduri_base + build_id
-        run.tag("BuildUri", value=build_uri)
-        run.parent.tag("BuildUri", value=build_uri)
     print(f"tags now present for run: {run.tags}")
 
     run.complete()

ml_service/pipelines/diabetes_regression_build_train_pipeline.py

@@ -38,15 +38,9 @@ def main():
 
     run_config = RunConfiguration()
     run_config.environment = environment
-    if (e.collection_uri is not None and e.teamproject_name is not None):
-        builduri_base = e.collection_uri + e.teamproject_name
-        builduri_base = builduri_base + "/_build/results?buildId="
-        run_config.environment.environment_variables["BUILDURI_BASE"] = builduri_base  # NOQA: E501
 
     model_name_param = PipelineParameter(
         name="model_name", default_value=e.model_name)
-    build_id_param = PipelineParameter(
-        name="build_id", default_value=e.build_id)
 
     # Get dataset name
     dataset_name = e.dataset_name
@@ -98,7 +92,6 @@ def main():
         inputs=[dataset.as_named_input('training_data')],
         outputs=[pipeline_data],
         arguments=[
-            "--build_id", build_id_param,
             "--model_name", model_name_param,
             "--step_output", pipeline_data
         ],
@@ -113,7 +106,6 @@ def main():
         compute_target=aml_compute,
         source_directory=e.sources_directory_train,
         arguments=[
-            "--build_id", build_id_param,
             "--model_name", model_name_param,
             "--allow_run_cancel", e.allow_run_cancel,
         ],
@@ -129,7 +121,6 @@ def main():
         source_directory=e.sources_directory_train,
         inputs=[pipeline_data],
         arguments=[
-            "--build_id", build_id_param,
             "--model_name", model_name_param,
             "--step_input", pipeline_data,
         ],

ml_service/pipelines/diabetes_regression_build_train_pipeline_with_r.py

@@ -34,10 +34,6 @@ def main():
         aml_workspace, e.aml_env_name, create_new=False)  # NOQA: E501
     run_config = RunConfiguration()
     run_config.environment = environment
-    if (e.collection_uri is not None and e.teamproject_name is not None):
-        builduri_base = e.collection_uri + e.teamproject_name
-        builduri_base = builduri_base + "/_build/results?buildId="
-        run_config.environment.environment_variables["BUILDURI_BASE"] = builduri_base  # NOQA: E501
 
     train_step = PythonScriptStep(
         name="Train Model",

ml_service/pipelines/run_train_pipeline.py

@@ -1,5 +1,5 @@
 from azureml.pipeline.core import PublishedPipeline
-from azureml.core import Workspace
+from azureml.core import Experiment, Workspace
 import argparse
 from ml_service.util.env_variables import Env
 
@@ -55,10 +55,16 @@ def main():
 
         if(args.skip_train_execution is False):
             pipeline_parameters = {"model_name": e.model_name}
-            run = published_pipeline.submit(
-                aml_workspace,
-                e.experiment_name,
-                pipeline_parameters)
+            tags = {"BuildId": e.build_id}
+            if (e.build_uri is not None):
+                tags["BuildUri"] = e.build_uri
+            experiment = Experiment(
+                workspace=aml_workspace,
+                name=e.experiment_name)
+            run = experiment.submit(
+                published_pipeline,
+                tags=tags,
+                pipeline_parameters=pipeline_parameters)
 
             print("Pipeline run initiated ", run.id)
 

ml_service/util/env_variables.py

@@ -38,8 +38,7 @@ def __init__(self):
         self._image_name = os.environ.get('IMAGE_NAME')
         self._db_cluster_id = os.environ.get("DB_CLUSTER_ID")
         self._score_script = os.environ.get("SCORE_SCRIPT")
-        self._collection_uri = os.environ.get("SYSTEM_COLLECTIONURI")
-        self._teamproject_name = os.environ.get("SYSTEM_TEAMPROJECT")
+        self._build_uri = os.environ.get("BUILD_URI")
         self._dataset_name = os.environ.get("DATASET_NAME")
         self._run_evaluation = os.environ.get("RUN_EVALUATION", "true")
         self._allow_run_cancel = os.environ.get(
@@ -139,12 +138,8 @@ def score_script(self):
         return self._score_script
 
     @property
-    def collection_uri(self):
-        return self._collection_uri
-
-    @property
-    def teamproject_name(self):
-        return self._teamproject_name
+    def build_uri(self):
+        return self._build_uri
 
     @property
     def dataset_name(self):