feat: ✨ Make nodes, leaves, depth labels customizable in .env

benchmark/Datasets.py

@@ -109,14 +109,14 @@ class DatasetsSurcov:
 
 class Datasets:
     def __init__(self, dataset_name=None, discretize=None):
-        envData = EnvData.load()
+        env_data = EnvData.load()
         # DatasetsSurcov, DatasetsTanveer, DatasetsArff,...
         source_name = getattr(
             __import__(__name__),
-            f"Datasets{envData['source_data']}",
+            f"Datasets{env_data['source_data']}",
         )
         self.discretize = (
-            envData["discretize"] == "1"
+            env_data["discretize"] == "1"
             if discretize is None
             else discretize == "1"
         )

benchmark/Results.py

@@ -65,6 +65,11 @@ class BaseReport(abc.ABC):
             self.lines = self.data["results"]
             self.score_name = self.data["score_name"]
         self.__compute_best_results_ever()
+        # Set the labels for nodes, leaves, depth
+        env_data = EnvData.load()
+        self.nodes_label = env_data["nodes"]
+        self.leaves_label = env_data["leaves"]
+        self.depth_label = env_data["depth"]
 
     def __compute_best_results_ever(self):
         args = EnvData.load()
@@ -132,23 +137,23 @@ class BaseReport(abc.ABC):
 
 class Report(BaseReport):
     header_lengths = [30, 6, 5, 3, 7, 7, 7, 15, 17, 15]
-    header_cols = [
-        "Dataset",
-        "Sampl.",
-        "Feat.",
-        "Cls",
-        "Nodes",
-        "Leaves",
-        "Depth",
-        "Score",
-        "Time",
-        "Hyperparameters",
-    ]
 
     def __init__(self, file_name: str, compare: bool = False):
         super().__init__(file_name)
         self.nline = 0
         self.compare = compare
+        self.header_cols = [
+            "Dataset",
+            "Sampl.",
+            "Feat.",
+            "Cls",
+            self.nodes_label,
+            self.leaves_label,
+            self.depth_label,
+            "Score",
+            "Time",
+            "Hyperparameters",
+        ]
 
     def header_line(self, text: str) -> None:
         print(TextColor.LINE1, end="")
@@ -489,9 +494,9 @@ class Excel(BaseReport):
             ("Samples", 10),
             ("Features", 7),
             ("Classes", 7),
-            ("Nodes", 7),
-            ("Leaves", 7),
-            ("Depth", 7),
+            (self.nodes_label, 7),
+            (self.leaves_label, 7),
+            (self.depth_label, 7),
             ("Score", 12),
             ("Score Std.", 12),
             ("Time", 12),

benchmark/tests/.env

@@ -6,4 +6,7 @@ stratified=0
 # Source of data Tanveer/Surcov
 source_data=Tanveer
 seeds=[57, 31, 1714, 17, 23, 79, 83, 97, 7, 1]
-discretize=0
+discretize=0
+nodes=Nodes
+leaves=Leaves
+depth=Depth

benchmark/tests/.env.arff

@@ -5,4 +5,7 @@ model=ODTE
 stratified=0
 source_data=Arff
 seeds=[271, 314, 171]
-discretize=1
+discretize=1
+nodes=Nodes
+leaves=Leaves
+depth=Depth

benchmark/tests/.env.dist

@@ -6,4 +6,7 @@ stratified=0
 # Source of data Tanveer/Surcov
 source_data=Tanveer
 seeds=[57, 31, 1714, 17, 23, 79, 83, 97, 7, 1]
-discretize=0
+discretize=0
+nodes=Nodes
+leaves=Leaves
+depth=Depth

benchmark/tests/.env.surcov

@@ -6,4 +6,7 @@ stratified=0
 # Source of data Tanveer/Surcov
 source_data=Surcov
 seeds=[57, 31, 1714, 17, 23, 79, 83, 97, 7, 1]
-discretize=0
+discretize=0
+nodes=Nodes
+leaves=Leaves
+depth=Depth

benchmark/tests/Util_test.py

@@ -180,6 +180,9 @@ class UtilTest(TestBase):
             "source_data": "Tanveer",
             "seeds": "[57, 31, 1714, 17, 23, 79, 83, 97, 7, 1]",
             "discretize": "0",
+            "nodes": "Nodes",
+            "leaves": "Leaves",
+            "depth": "Depth",
         }
         computed = EnvData().load()
         self.assertDictEqual(computed, expected)