@@ -85,7 +85,11 @@ def evaluate(
                 continue
         benchmark = benchmark_cls()
         for name in benchmark.names():
-            if Result(method_name=name, benchmark_name=benchmark_cls.name) in results:
+            if Result(
+                code=benchmark.code,
+                method_name=name,
+                benchmark_name=benchmark_cls.name,
+            ) in results:
                 continue
             else:
                 result = benchmark.evaluate(name, on_error=on_error)
 
@@ -59,7 +59,7 @@ def get_predictions(self):
         raise NotImplementedError()
 
     def __init__(self):
-        keys = ["name", "keypoints", "ground_truth", "metadata"]
+        keys = ["code", "name", "keypoints", "ground_truth", "metadata"]
         for key in keys:
             if not hasattr(self, key):
                 raise NotImplementedError(
@@ -110,6 +110,7 @@ def evaluate(self, name: str, on_error="raise"):
             else:
                 raise NotImplementedError() from exception
         return Result(
+            code=self.code,
             method_name=name,
             benchmark_name=self.name,
             mean_avg_precision=mean_avg_precision,
@@ -139,13 +140,15 @@ def _validate_predictions(self, name: str, predictions: dict) -> dict:
 class Result:
     """Benchmark result."""
 
+    code: str
     method_name: str
     benchmark_name: str
     root_mean_squared_error: float = float("nan")
     mean_avg_precision: float = float("nan")
     benchmark_version: str = __version__
 
     _export_mapping = dict(
+        code="code",
         benchmark_name="benchmark",
         method_name="method",
         benchmark_version="version",
 
@@ -96,6 +96,16 @@ def compute_pose_map(self, results_objects):
             symmetric_kpts=[(0, 4), (1, 3)],
         )
 
+    def _validate_predictions(self, name: str, predictions: dict) -> dict:
+        """Fixes filenames for predictions made on old versions of the dataset"""
+        return super()._validate_predictions(
+            name,
+            {
+                k.replace("Dummy", "D").replace("Dead pup", "DP"): v
+                for k, v in predictions.items()
+            },
+        )
+
 
 class MarmosetBenchmark(deeplabcut.benchmark.base.Benchmark):
     """Dataset with two marmosets.