diff --git a/ScaFFold/worker.py b/ScaFFold/worker.py
index 61b1c41..a51c51e 100644
--- a/ScaFFold/worker.py
+++ b/ScaFFold/worker.py
@@ -284,8 +284,6 @@ def main(kwargs_dict: dict = {}):
         outfile_path = trainer.outfile_path
         train_data = np.genfromtxt(outfile_path, dtype=float, delimiter=",", names=True)
         total_train_time = train_data["epoch_duration"].sum()
-        fom = 1.0 / total_train_time
-        adiak_value("FOM", fom)
         if "total_optimizer_steps" in train_data.dtype.names:
             optimizer_steps = np.atleast_1d(train_data["total_optimizer_steps"])
             total_optimizer_steps = int(optimizer_steps[-1])
@@ -296,15 +294,16 @@ def main(kwargs_dict: dict = {}):
         else:
             total_optimizer_steps = int(getattr(trainer, "total_optimizer_steps", 0))
         adiak_value("total_optimizer_steps", total_optimizer_steps)
-        log.info(
-            f"FOM = {fom} (1 / total_train_time={total_train_time:.6f} seconds). "
-            f"This FOM is specific to problem_scale={config.problem_scale}, "
-            f"target_dice={config.target_dice}, seed={config.seed}, "
-            f"total_optimizer_steps={total_optimizer_steps}."
-        )
         epochs = np.atleast_1d(train_data["epoch"])
         total_epochs = int(epochs[-1])
         if config.epochs == -1:
+            fom = 1.0 / total_train_time
+            adiak_value("FOM", fom)
+            log.info(
+                f"FOM = {fom} (1 / total_train_time={total_train_time:.6f} seconds). "
+                f"This FOM is specific to problem_scale={config.problem_scale}, "
+                f"target_dice={config.target_dice}, seed={config.seed}."
+            )
             extra_msg = f"Trained to >= {config.target_dice} validation dice score in {total_train_time:.2f} seconds, {total_epochs} epochs, {total_optimizer_steps} optimizer steps."
         else:
             extra_msg = f"Completed in {total_train_time:.2f} seconds, {total_epochs} epochs, {total_optimizer_steps} optimizer steps."