Refactor Object constructor and include eta_range as obj_key postfix

menu_tools/object_performance/config.py

@@ -53,12 +53,15 @@ def test_objects(self) -> dict[str, Any]:
         if not all([":" in x for x in self._cfg["test_objects"]]):
             raise ValueError(f"Misconfigured obj:id key in {self.plot_name}!")
 
-        test_obj = {
-            x: {"base_obj": x.split(":")[0], "id": x.split(":")[1], "x_arg": x_arg}
-            for x, x_arg in self._cfg["test_objects"].items()
-        }
+        return self._cfg["test_objects"]
 
-        return test_obj
+        # DEPRECATED
+        # test_obj = {
+        #     x: {"base_obj": x.split(":")[0], "id": x.split(":")[1], "x_arg": x_arg}
+        #     for x, x_arg in self._cfg["test_objects"].items()
+        # }
+
+        # return test_obj
 
     @property
     def matching(self):
@@ -95,8 +98,6 @@ def ylabel(self):
     @property
     def test_object_instances(self) -> list:
         test_objects = []
-        for obj in self._cfg["test_objects"]:
-            nano_obj_name = obj.split(":")[0]
-            obj_id_name = obj.split(":")[1]
-            test_objects.append(Object(nano_obj_name, obj_id_name, self.version))
+        for obj_key in self._cfg["test_objects"]:
+            test_objects.append(Object(obj_key, self.version))
         return test_objects

menu_tools/object_performance/plotter.py

@@ -81,8 +81,7 @@ def _save_json(self, file_name):
             if obj_key == "ref":
                 continue
             obj = Object(
-                nano_obj_name=obj_key.split("_")[0],
-                obj_id_name=obj_key.split("_")[1],
+                obj_key,
                 version=self.version,
             )
 
@@ -149,8 +148,7 @@ def _plot_efficiency_curve(self):
             efficiency, yerr = self.turnon_collection.get_efficiency(obj_key)
 
             obj = Object(
-                nano_obj_name=obj_key.split("_")[0],
-                obj_id_name=obj_key.split("_")[1],
+                obj_key,
                 version=self.version,
             )
 
@@ -196,8 +194,7 @@ def _plot_iso_vs_efficiency_curve(self):
             iso_vs_eff_hist = self._get_iso_vs_eff_hist(gen_hist_trig[0])
 
             obj = Object(
-                nano_obj_name=obj_key.split("_")[0],
-                obj_id_name=obj_key.split("_")[1],
+                obj_key,
                 version=self.version,
             )
 

menu_tools/object_performance/tests/test_electron_v29.py

@@ -35,8 +35,10 @@ def test_isolation_barrel():
 
     for key, val in reference_data.items():
         if isinstance(val, dict):
+            if "tkEle" in key:
+                test_key = "tkElectron:NoIso:inclusive"
             efficiencies_test = np.array(
-                test_result[key]["efficiency"], dtype=np.float64
+                test_result[test_key]["efficiency"], dtype=np.float64
             )
             efficiencies_reference = np.array(val["efficiency"], dtype=np.float64)
             print(efficiencies_reference)

menu_tools/object_performance/turnon_collection.py

@@ -63,9 +63,7 @@ def _load_test_branches(self) -> None:
         """
         Load test objects.
         """
-        test_objects = self.cfg_plot.test_objects
-        for test_obj, obj_cfg in test_objects.items():
-            obj = Object(obj_cfg["base_obj"], obj_cfg["id"], self.cfg_plot.version)
+        for obj in self.cfg_plot.test_object_instances:
             test_array = self._load_array_from_parquet(obj.nano_obj_name)
             test_array = ak.with_name(test_array, "Momentum4D")
             self.turnon_collection.ak_arrays[str(obj)] = test_array
@@ -99,10 +97,9 @@ def test_objects(self) -> list[tuple[Object, str]]:
         obj_args = []
 
         test_objects = self.cfg_plot.test_objects
-        for test_obj, obj_cfg in test_objects.items():
-            obj = Object(obj_cfg["base_obj"], obj_cfg["id"], self.cfg_plot.version)
-            x_arg = obj_cfg["x_arg"].lower()
-            obj_args.append((obj, x_arg))
+        for obj_key, x_arg in test_objects.items():
+            obj = Object(obj_key, self.cfg_plot.version)
+            obj_args.append((obj, x_arg.lower()))
 
         return obj_args
 

menu_tools/rate_plots/config.py

@@ -44,10 +44,8 @@ def test_objects(self) -> list:
     @property
     def test_object_instances(self) -> dict[str, dict[str, Object]]:
         test_objects: dict[str, dict[str, Object]] = {}
-        for obj in self._cfg["test_objects"]:
-            nano_obj_name = obj.split(":")[0]
-            obj_id_name = obj.split(":")[1]
-            test_objects[obj] = {}
+        for obj_key in self._cfg["test_objects"]:
+            test_objects[obj_key] = {}
             for version in self.versions:
-                test_objects[obj][version] = Object(nano_obj_name, obj_id_name, version)
+                test_objects[obj_key][version] = Object(obj_key, version)
         return test_objects

menu_tools/utils/objects.py

@@ -17,22 +17,50 @@ class Object:
         version: version of the menu
     """
 
-    def __init__(self, nano_obj_name: str, obj_id_name: str, version: str) -> None:
+    def __init__(
+        self,
+        object_key: str,
+        version: str,
+    ) -> None:
         """Initializes an Object loading the parameters from the
            corresponding config file.
 
         Args:
-            nano_obj_name: name of the physics object in the l1 ntuples
-            obj_id_name: name of the l1 object id as defined in `configs`
+            object_key: object/id specifier of the form l1_object:id[:eta_range]
             version: version of the menu
         """
-        self.nano_obj_name = nano_obj_name
-        self.obj_id_name = obj_id_name
+        self.object_key = object_key
         self.version = version
         self._nano_obj  # fail early if no config can be found
 
     def __str__(self) -> str:
-        return f"{self.nano_obj_name}_{self.obj_id_name}"
+        return f"{self.nano_obj_name}:{self.obj_id_name}:{self.eta_range}"
+
+    @property
+    def file_ext(self) -> str:
+        return str(self).replace(":", "_")
+
+    @property
+    def nano_obj_name(self) -> str:
+        return self.object_key.split(":")[0]
+
+    @property
+    def obj_id_name(self) -> str:
+        return self.object_key.split(":")[1]
+
+    @property
+    def eta_range(self) -> str:
+        """If an eta range other than "inclusive" is specified, a cut to that
+        range is added to `cuts`.
+
+        Returns:
+            eta_range_key: `barrel`/`endcap`/`overlap`/`forward`/`inclusive`
+        """
+        try:
+            eta_range_key = self.object_key.split(":")[2]
+        except IndexError:
+            eta_range_key = "inclusive"
+        return eta_range_key
 
     @property
     def _nano_obj(self) -> dict[str, dict]:
@@ -147,8 +175,8 @@ def compute_selection_mask_for_object_cuts(obj: Object, ak_array: ak.Array) -> a
 
 
 if __name__ == "__main__":
-    x = Object("tkElectron", "Iso", "V29")
-    x = Object("caloJet", "default", "V29")
+    x = Object("tkElectron:Iso", "V29")
+    x = Object("caloJet:default", "V29")
     print(x)
     print(x.match_dR)
     print(x.plot_label)