update TF child function

Author: NihalHarish

src/sagemaker/estimator.py

@@ -2219,6 +2219,7 @@ def _validate_and_set_debugger_configs(self):
         ):
             self.debugger_hook_config = DebuggerHookConfig(s3_output_path=self.output_path)
         elif not self.debugger_hook_config:
+            # set hook config to False if _region_supports_debugger is False
             self.debugger_hook_config = False
 
         # Disable debugger if checkpointing is enabled by the customer

src/sagemaker/tensorflow/estimator.py

@@ -18,7 +18,6 @@
 from packaging import version
 
 from sagemaker import image_uris, s3, utils
-from sagemaker.debugger import DebuggerHookConfig
 from sagemaker.deprecations import renamed_kwargs
 from sagemaker.estimator import Framework
 import sagemaker.fw_utils as fw
@@ -347,6 +346,7 @@ def _validate_and_set_debugger_configs(self):
 
         Else, set default HookConfig
         """
+        super(TensorFlow, self)._validate_and_set_debugger_configs()
         ps_enabled = "parameter_server" in self.distribution and self.distribution[
             "parameter_server"
         ].get("enabled", False)
@@ -358,11 +358,6 @@ def _validate_and_set_debugger_configs(self):
                 )
             self.debugger_hook_config = None
             self.debugger_rule_configs = None
-        elif self.debugger_hook_config is None and fw._region_supports_debugger(
-            self.sagemaker_session.boto_session.region_name
-        ):
-            # Set defaults for debugging.
-            self.debugger_hook_config = DebuggerHookConfig(s3_output_path=self.output_path)
 
     def transformer(
         self,

tests/integ/test_debugger.py

@@ -25,6 +25,7 @@
 )
 from sagemaker.mxnet.estimator import MXNet
 from sagemaker.pytorch.estimator import PyTorch
+from sagemaker.tensorflow.estimator import TensorFlow
 from sagemaker.xgboost.estimator import XGBoost
 from tests.integ import DATA_DIR, TRAINING_DEFAULT_TIMEOUT_MINUTES
 from tests.integ.retry import retries
@@ -424,6 +425,26 @@ def test_debug_hook_disabled_with_checkpointing(
         # Debug Hook should be disabled
         assert pt.debugger_hook_config is False
 
+        # Estimator with checkpointing enabled and SMModelParallel Enabled
+        tf = TensorFlow(
+            base_job_name="tf-smdataparallel-mnist",
+            entry_point=script_path,
+            role="SageMakerRole",
+            framework_version="2.3.1",
+            py_version="py36",
+            instance_count=1,
+            # For training with p3dn instance use - ml.p3dn.24xlarge, with p4dn instance use - ml.p4d.24xlarge
+            instance_type="ml.p3.16xlarge",
+            sagemaker_session=sagemaker_session,
+            # Training using SMDataParallel Distributed Training Framework
+            distribution={"smdistributed": {"modelparallel": {"enabled": True}}},
+            checkpoint_local_path="/opt/ml/checkpoints",
+            checkpoint_s3_uri=os.path.join(s3_output_path, "checkpoints"),
+        )
+        tf._prepare_for_training()
+        # Debug Hook should be disabled
+        assert tf.debugger_hook_config is False
+
         # Estimator with checkpointing enabled with Xgboost Estimator
         xg = XGBoost(
             base_job_name="test_xgboost",