Fix for Estimator Training details on register

Author: selvask-aws

src/sagemaker/estimator.py

@@ -68,6 +68,7 @@
 from sagemaker.interactive_apps import SupportedInteractiveAppTypes
 from sagemaker.interactive_apps.tensorboard import TensorBoardApp
 from sagemaker.instance_group import InstanceGroup
+from sagemaker.model_card.model_card import ModelCard, TrainingDetails
 from sagemaker.utils import instance_supports_kms
 from sagemaker.job import _Job
 from sagemaker.jumpstart.utils import (
@@ -1797,8 +1798,16 @@ def register(
         else:
             if "model_kms_key" not in kwargs:
                 kwargs["model_kms_key"] = self.output_kms_key
-            model = self.create_model(image_uri=image_uri, **kwargs)
+            model = self.create_model(image_uri=image_uri, name=model_name, **kwargs)
         model.name = model_name
+        if self.model_data is not None and model_card is None:
+            training_details = TrainingDetails.from_model_s3_artifacts(
+                model_artifacts=[self.model_data], sagemaker_session=self.sagemaker_session
+            )
+            model_card = ModelCard(
+                name="estimator_card",
+                training_details=training_details,
+            )
         return model.register(
             content_types,
             response_types,

src/sagemaker/model.py

@@ -549,7 +549,10 @@ def register(
             model_package_group_name = utils.base_name_from_image(
                 self.image_uri, default_base_name=ModelPackage.__name__
             )
-        if model_package_group_name is not None:
+        if (
+            model_package_group_name is not None
+            and self.model_type is not JumpStartModelType.PROPRIETARY
+        ):
             container_def = self.prepare_container_def(accept_eula=accept_eula)
             container_def = update_container_with_inference_params(
                 framework=framework,

tests/integ/test_byo_estimator.py

@@ -12,14 +12,20 @@
 # language governing permissions and limitations under the License.
 from __future__ import absolute_import
 
+import io
 import json
 import os
 
+import numpy as np
+
 import pytest
+import sagemaker.amazon.common as smac
+
 
 import sagemaker
 from sagemaker import image_uris
 from sagemaker.estimator import Estimator
+from sagemaker.s3 import S3Uploader
 from sagemaker.serializers import SimpleBaseSerializer
 from sagemaker.utils import unique_name_from_base
 from tests.integ import DATA_DIR, TRAINING_DEFAULT_TIMEOUT_MINUTES, datasets
@@ -102,6 +108,61 @@ def test_byo_estimator(sagemaker_session, region, cpu_instance_type, training_se
             assert prediction["score"] is not None
 
 
+def test_estimator_register_publish_training_details(
+    sagemaker_session, region, cpu_instance_type, training_set
+):
+
+    bucket = sagemaker_session.default_bucket()
+    prefix = "model-card-sample-notebook"
+
+    raw_data = (
+        (0.5, 0),
+        (0.75, 0),
+        (1.0, 0),
+        (1.25, 0),
+        (1.50, 0),
+        (1.75, 0),
+        (2.0, 0),
+        (2.25, 1),
+        (2.5, 0),
+        (2.75, 1),
+        (3.0, 0),
+        (3.25, 1),
+        (3.5, 0),
+        (4.0, 1),
+        (4.25, 1),
+        (4.5, 1),
+        (4.75, 1),
+        (5.0, 1),
+        (5.5, 1),
+    )
+    training_data = np.array(raw_data).astype("float32")
+    labels = training_data[:, 1]
+
+    # upload data to S3 bucket
+    buf = io.BytesIO()
+    smac.write_numpy_to_dense_tensor(buf, training_data, labels)
+    buf.seek(0)
+    s3_train_data = f"s3://{bucket}/{prefix}/train"
+    S3Uploader.upload_bytes(b=buf, s3_uri=s3_train_data, sagemaker_session=sagemaker_session)
+    output_location = f"s3://{bucket}/{prefix}/output"
+    container = image_uris.retrieve("linear-learner", region)
+    estimator = sagemaker.estimator.Estimator(
+        container,
+        role="SageMakerRole",
+        instance_count=1,
+        instance_type="ml.m4.xlarge",
+        output_path=output_location,
+        sagemaker_session=sagemaker_session,
+    )
+    estimator.set_hyperparameters(
+        feature_dim=2, mini_batch_size=10, predictor_type="binary_classifier"
+    )
+    estimator.fit({"train": s3_train_data})
+    print(f"Training job name: {estimator.latest_training_job.name}")
+    estimator.register()
+
+
 def test_async_byo_estimator(sagemaker_session, region, cpu_instance_type, training_set):
     image_uri = image_uris.retrieve("factorization-machines", region)
     endpoint_name = unique_name_from_base("byo")