fix(train): accept ModelTrainer Tag objects when creating the tuning job

sagemaker-train/src/sagemaker/train/tuner.py

@@ -1277,7 +1277,18 @@ def _start_tuning_job(self, inputs):
         if self.tags:
             from sagemaker.core.shapes import Tag
 
-            tag_objects = [Tag(key=tag["Key"], value=tag["Value"]) for tag in self.tags]
+            # self.tags mixes Key/Value dicts (tuner/JumpStart tags) with the Tag objects
+            # merged in from ModelTrainer.tags.
+            tag_objects = [
+                (
+                    tag
+                    if isinstance(tag, Tag)
+                    else Tag(
+                        key=tag.get("Key", tag.get("key")), value=tag.get("Value", tag.get("value"))
+                    )
+                )
+                for tag in self.tags
+            ]
 
         # Build tuning request
         tuning_request = {

sagemaker-train/tests/unit/train/test_tuner_tags.py

@@ -0,0 +1,68 @@
+# Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"). You
+# may not use this file except in compliance with the License. A copy of
+# the License is located at
+#
+#     http://aws.amazon.com/apache2.0/
+#
+# or in the "license" file accompanying this file. This file is
+# distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF
+# ANY KIND, either express or implied. See the License for the specific
+# language governing permissions and limitations under the License.
+"""Tags passed to the HyperParameterTuningJob created by HyperparameterTuner."""
+
+from __future__ import absolute_import
+
+from unittest.mock import MagicMock, patch
+
+import pytest
+
+from sagemaker.core.helper.session_helper import Session
+from sagemaker.core.parameter import ContinuousParameter
+from sagemaker.core.shapes import Tag
+from sagemaker.train.configs import Compute
+from sagemaker.train.model_trainer import ModelTrainer
+from sagemaker.train.tuner import HyperparameterTuner
+
+ROLE = "arn:aws:iam::123456789012:role/role"
+IMAGE = "123456789012.dkr.ecr.us-west-2.amazonaws.com/image:latest"
+
+
+@pytest.fixture
+def session():
+    session = MagicMock(spec=Session)
+    session.default_bucket.return_value = "bucket"
+    session.default_bucket_prefix = None
+    session.boto_region_name = "us-west-2"
+    session.boto_session = MagicMock()
+    session.sagemaker_config = None
+    session.settings = MagicMock(include_jumpstart_tags=False)
+    return session
+
+
+@patch("sagemaker.train.tuner.HyperParameterTuningJob.create")
+@patch("sagemaker.train.defaults.resolve_and_validate_role", return_value=ROLE)
+def test_tune_accepts_model_trainer_tag_objects_and_tag_dicts(_, mock_create, session):
+    model_trainer = ModelTrainer(
+        training_image=IMAGE,
+        role=ROLE,
+        sagemaker_session=session,
+        compute=Compute(instance_type="ml.m5.xlarge", instance_count=1),
+        hyperparameters={"lr": 0.1},
+        tags=[Tag(key="team", value="ml")],
+    )
+    tuner = HyperparameterTuner(
+        model_trainer,
+        "accuracy",
+        {"lr": ContinuousParameter(0.001, 0.1)},
+        max_jobs=2,
+        tags=[{"Key": "project", "Value": "hpo"}],
+    )
+
+    tuner.tune(inputs="s3://bucket/train", wait=False)
+
+    assert mock_create.call_args.kwargs["tags"] == [
+        Tag(key="project", value="hpo"),
+        Tag(key="team", value="ml"),
+    ]