Rename trial_timeout to trial_active_deadline_seconds

- Updated parameter name in function signatures and documentation
- Updated test cases and assertions to use new parameter name
- Fixed line length issues to meet linting requirements
- Maintains backward compatibility with Kubernetes activeDeadlineSeconds field

Author: wassimbensalem

sdk/python/v1beta1/kubeflow/katib/api/katib_client.py

@@ -216,7 +216,7 @@ def tune(
         packages_to_install: List[str] = None,
         pip_index_url: str = "https://pypi.org/simple",
         metrics_collector_config: Dict[str, Any] = {"kind": "StdOut"},
-        trial_timeout: Optional[int] = None,
+        trial_active_deadline_seconds: Optional[int] = None,
     ):
         """
         Create HyperParameter Tuning Katib Experiment using one of the following
@@ -356,7 +356,8 @@ class name in this argument.
             metrics_collector_config: Specify the config of metrics collector,
                 for example, `metrics_collector_config = {"kind": "Push"}`.
                 Currently, we only support `StdOut` and `Push` metrics collector.
-            trial_timeout: Optional timeout in seconds for each trial. If None, no timeout is
+            trial_active_deadline_seconds: Optional timeout in seconds for each trial.
+                If None, no timeout is
                 applied. For Job-based trials, this sets the activeDeadlineSeconds field.
                 For PyTorchJob-based trials, this sets the activeDeadlineSeconds field on the
                 Master replica. This prevents individual trials from running indefinitely and
@@ -517,15 +518,15 @@ class name in this argument.
                     resources_per_trial,
                     training_utils.get_pod_template_spec(containers=[container_spec]),
                     training_utils.get_pod_template_spec(containers=[container_spec]),
-                    trial_timeout,
+                    trial_active_deadline_seconds,
                 )
             # Otherwise, Trial uses Job for model training.
             else:
                 trial_template = utils.get_trial_template_with_job(
                     retain_trials,
                     trial_parameters,
                     training_utils.get_pod_template_spec(containers=[container_spec]),
-                    trial_timeout,
+                    trial_active_deadline_seconds,
                 )
 
         # If users choose to use external models and datasets.
@@ -702,7 +703,7 @@ class name in this argument.
                 resources_per_trial,
                 master_pod_template_spec,
                 worker_pod_template_spec,
-                trial_timeout,
+                trial_active_deadline_seconds,
             )
 
         # Add parameters to the Katib Experiment.

sdk/python/v1beta1/kubeflow/katib/api/katib_client_test.py

@@ -542,19 +542,19 @@ def create_experiment(
         TEST_RESULT_SUCCESS,
     ),
     (
-        "valid flow with trial_timeout for Job-based trials",
+        "valid flow with trial_active_deadline_seconds for Job-based trials",
         {
             "name": "tune_test",
             "objective": lambda x: print(f"a={x}"),
             "parameters": {"a": katib.search.int(min=10, max=100)},
             "objective_metric_name": "a",
             "resources_per_trial": {"gpu": "2"},
-            "trial_timeout": 3600,  # 1 hour timeout
+            "trial_active_deadline_seconds": 3600,  # 1 hour timeout
         },
         TEST_RESULT_SUCCESS,
     ),
     (
-        "valid flow with trial_timeout for PyTorchJob-based trials",
+        "valid flow with trial_active_deadline_seconds for PyTorchJob-based trials",
         {
             "name": "tune_test",
             "objective": lambda x: print(f"a={x}"),
@@ -565,7 +565,7 @@ def create_experiment(
                 num_procs_per_worker=2,
                 resources_per_worker={"gpu": "2"},
             ),
-            "trial_timeout": 7200,  # 2 hours timeout
+            "trial_active_deadline_seconds": 7200,  # 2 hours timeout
         },
         TEST_RESULT_SUCCESS,
     ),
@@ -695,8 +695,10 @@ def test_tune(katib_client, test_name, kwargs, expected_output):
                         KubeflowOrgV1PyTorchJob,
                     )
 
-                elif test_name == "valid flow with trial_timeout for Job-based trials":
-                    # Verify trial_timeout is set on Job spec
+                elif test_name == (
+                    "valid flow with trial_active_deadline_seconds for Job-based trials"
+                ):
+                    # Verify trial_active_deadline_seconds is set on Job spec
                     job_spec = experiment.spec.trial_template.trial_spec.spec
                     assert job_spec.active_deadline_seconds == 3600
                     # Verify other Job-specific fields
@@ -710,9 +712,9 @@ def test_tune(katib_client, test_name, kwargs, expected_output):
 
                 elif (
                     test_name
-                    == "valid flow with trial_timeout for PyTorchJob-based trials"
+                    == "valid flow with trial_active_deadline_seconds for PyTorchJob-based trials"
                 ):
-                    # Verify trial_timeout is set on PyTorchJob run_policy
+                    # Verify trial_active_deadline_seconds is set on PyTorchJob run_policy
                     pytorch_spec = experiment.spec.trial_template.trial_spec.spec
                     assert pytorch_spec.run_policy.active_deadline_seconds == 7200
                     # Verify PyTorchJob type

sdk/python/v1beta1/kubeflow/katib/utils/utils.py

@@ -284,7 +284,7 @@ def get_trial_template_with_job(
     retain_trials: bool,
     trial_parameters: List[models.V1beta1TrialParameterSpec],
     pod_template_spec: client.V1PodTemplateSpec,
-    trial_timeout: Optional[int] = None,
+    trial_active_deadline_seconds: Optional[int] = None,
 ) -> models.V1beta1TrialTemplate:
     """
     Get Trial template with Job as a Trial's Worker
@@ -298,7 +298,7 @@ def get_trial_template_with_job(
         api_version="batch/v1",
         kind="Job",
         spec=client.V1JobSpec(
-            template=pod_template_spec, active_deadline_seconds=trial_timeout
+            template=pod_template_spec, active_deadline_seconds=trial_active_deadline_seconds
         ),
     )
 
@@ -317,7 +317,7 @@ def get_trial_template_with_pytorchjob(
     resources_per_trial: types.TrainerResources,
     master_pod_template_spec: models.V1PodTemplateSpec,
     worker_pod_template_spec: models.V1PodTemplateSpec,
-    trial_timeout: Optional[int] = None,
+    trial_active_deadline_seconds: Optional[int] = None,
 ) -> models.V1beta1TrialTemplate:
     """
     Get Trial template with PyTorchJob as a Trial's Worker
@@ -330,7 +330,7 @@ def get_trial_template_with_pytorchjob(
         spec=training_models.KubeflowOrgV1PyTorchJobSpec(
             run_policy=training_models.KubeflowOrgV1RunPolicy(
                 clean_pod_policy=None,
-                active_deadline_seconds=trial_timeout,
+                active_deadline_seconds=trial_active_deadline_seconds,
             ),
             nproc_per_node=str(resources_per_trial.num_procs_per_worker),
             pytorch_replica_specs={