[GitHub] [beam] damccorm commented on a diff in pull request #26261: DLQ support in RunInference

["damccorm (via GitHub)" <gi...@apache.org>]

damccorm commented on code in PR #26261:
URL: https://github.com/apache/beam/pull/26261#discussion_r1165888266


##########
sdks/python/apache_beam/ml/inference/base.py:
##########
@@ -368,20 +369,71 @@ def expand(
         # batching DoFn APIs.
         | beam.BatchElements(**self._model_handler.batch_elements_kwargs()))
 
+    run_inference_pardo = beam.ParDo(
+        _RunInferenceDoFn(
+            self._model_handler,
+            self._clock,
+            self._metrics_namespace,
+            self._enable_side_input_loading),
+        self._inference_args,
+        beam.pvalue.AsSingleton(
+            self._model_metadata_pcoll,
+        ) if self._enable_side_input_loading else None).with_resource_hints(
+            **resource_hints)
+
+    if self._with_exception_handling:
+      run_inference_pardo = run_inference_pardo.with_exception_handling(
+          exc_class=self._exc_class,
+          use_subprocess=self._use_subprocess,
+          threshold=self._threshold)
+
     return (
         batched_elements_pcoll
-        | 'BeamML_RunInference' >> (
-            beam.ParDo(
-                _RunInferenceDoFn(
-                    self._model_handler,
-                    self._clock,
-                    self._metrics_namespace,
-                    self._enable_side_input_loading),
-                self._inference_args,
-                beam.pvalue.AsSingleton(
-                    self._model_metadata_pcoll,
-                ) if self._enable_side_input_loading else
-                None).with_resource_hints(**resource_hints)))
+        | 'BeamML_RunInference' >> run_inference_pardo)
+
+  def with_exception_handling(
+      self, *, exc_class=Exception, use_subprocess=False, threshold=1):

Review Comment:
   I originally had it, but intentionally omitted it since those are primarily useful in the context of having >2 outputs



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@beam.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org