diff --git a/numalogic/udfs/trainer/_base.py b/numalogic/udfs/trainer/_base.py
index a12640d7..9ff27eff 100644
--- a/numalogic/udfs/trainer/_base.py
+++ b/numalogic/udfs/trainer/_base.py
@@ -219,7 +219,7 @@ def exec(self, keys: list[str], datum: Datum) -> Messages:
 
         # Construct feature array
         x_train, nan_counter, inf_counter = self.get_feature_arr(
-            df, payload.metrics, max_value_map=_conf.numalogic_conf.trainer.max_value_map
+            df, _conf.metrics, max_value_map=_conf.numalogic_conf.trainer.max_value_map
         )
         _add_summary(
             summary=NAN_SUMMARY,
diff --git a/tests/udfs/test_trainer.py b/tests/udfs/test_trainer.py
index 195ab042..6d9dd2ee 100644
--- a/tests/udfs/test_trainer.py
+++ b/tests/udfs/test_trainer.py
@@ -87,7 +87,9 @@ def test_trainer_01(self):
                                 conf={"seq_len": 12, "n_features": 2},
                             ),
                             preprocess=[ModelInfo(name="LogTransformer", stateful=True, conf={})],
-                            trainer=TrainerConf(pltrainer_conf=LightningTrainerConf(max_epochs=1)),
+                            trainer=TrainerConf(
+                                pltrainer_conf=LightningTrainerConf(accelerator="cpu", max_epochs=1)
+                            ),
                         ),
                     )
                 }
@@ -122,6 +124,7 @@ def test_trainer_03(self):
                 ml_pipelines={
                     "pipeline1": MLPipelineConf(
                         pipeline_id="pipeline1",
+                        metrics=["failed", "degraded"],
                         numalogic_conf=NumalogicConf(
                             model=ModelInfo(
                                 name="VanillaAE", conf={"seq_len": 12, "n_features": 2}
@@ -154,6 +157,7 @@ def test_trainer_do_train(self):
                 ml_pipelines={
                     "pipeline1": MLPipelineConf(
                         pipeline_id="pipeline1",
+                        metrics=["failed", "degraded"],
                         numalogic_conf=NumalogicConf(
                             model=ModelInfo(
                                 name="VanillaAE", conf={"seq_len": 12, "n_features": 2}
@@ -197,6 +201,7 @@ def test_trainer_do_not_train_1(self):
                 ml_pipelines={
                     "pipeline1": MLPipelineConf(
                         pipeline_id="pipeline1",
+                        metrics=["failed", "degraded"],
                         numalogic_conf=NumalogicConf(
                             model=ModelInfo(
                                 name="VanillaAE", conf={"seq_len": 12, "n_features": 2}
@@ -238,6 +243,7 @@ def test_trainer_do_not_train_2(self):
                 ml_pipelines={
                     "pipeline1": MLPipelineConf(
                         pipeline_id="pipeline1",
+                        metrics=["failed", "degraded"],
                         numalogic_conf=NumalogicConf(
                             model=ModelInfo(
                                 name="VanillaAE", conf={"seq_len": 12, "n_features": 2}
@@ -402,6 +408,7 @@ def test_trainer_datafetcher_err_and_train(self):
                     ml_pipelines={
                         "pipeline1": MLPipelineConf(
                             pipeline_id="pipeline1",
+                            metrics=["failed", "degraded"],
                             numalogic_conf=NumalogicConf(
                                 model=ModelInfo(
                                     name="VanillaAE", conf={"seq_len": 12, "n_features": 2}