Add timer

main.py

@@ -148,6 +148,9 @@ if __name__ == "__main__":
             with open("algorithm_params.json", "r", encoding="utf-8") as f:
                 algo_params = json.load(f)
 
+            with open(f"{INPUT_DIR}/{config_name}/time.csv", "a", encoding="utf-8") as f:
+                f.write("Algorithm,dataset,duration")
+
             for algo in args.algorithms:
                 params = algo_params[algo]
                 train = params.pop("training")

split/base.py

@@ -9,6 +9,8 @@ import numpy as np
 import pandas as pd
 import seaborn as sns
 
+from time import time
+
 from sklearn.cluster import KMeans, HDBSCAN
 from sklearn.exceptions import ConvergenceWarning
 from sklearn.metrics import silhouette_score
@@ -79,19 +81,26 @@ class BaseSplitter:
         """
         Compute the vector of correlation coefficients for each of the variable of the dataset.
         """
-        x = None
-        for coeff in CORRELATION_CLASSES:
-            coeff_name = coeff.__name__[:-11]
-            correlation_matrix = coeff().compute(data)
-            correlation_matrix.to_csv(f"{self.data_path}/dataset_correlation_{coeff_name}.csv")
-            sns.heatmap(correlation_matrix, annot=True, cmap="coolwarm")\
-                .get_figure()\
-                .savefig(f"{self.data_path}/dataset_correlation_matrix_{coeff_name}.png")
-            plt.clf()
-
-            if x is not None:
-                x = np.concatenate([np.abs(correlation_matrix), x], axis=1)
-            else:
-                x = np.abs(correlation_matrix)
+        with open(f"{self.data_path}/split_time.csv", "a", encoding="utf-8") as f:
+            f.write("Algorithm,duration")
+            x = None
+            for coeff in CORRELATION_CLASSES:
+                coeff_name = coeff.__name__[:-11]
+
+                start_time = time()
+                correlation_matrix = coeff().compute(data)
+                duration = time() - start_time
+                f.write(f"{coeff_name},{duration}")
+
+                correlation_matrix.to_csv(f"{self.data_path}/dataset_correlation_{coeff_name}.csv")
+                sns.heatmap(correlation_matrix, annot=True, cmap="coolwarm")\
+                    .get_figure()\
+                    .savefig(f"{self.data_path}/dataset_correlation_matrix_{coeff_name}.png")
+                plt.clf()
+
+                if x is not None:
+                    x = np.concatenate([np.abs(correlation_matrix), x], axis=1)
+                else:
+                    x = np.abs(correlation_matrix)
 
         return x

trainers/base.py

@@ -7,6 +7,7 @@ import pandas as pd
 import seaborn as sns
 import subprocess
 
+from time import time
 
 from .correlations import CORRELATION_CLASSES
 
@@ -58,6 +59,11 @@ class BaseTrainer():
 
             os.makedirs(f"{path}/results_{self.algorithm}", exist_ok=True)
 
+            start_time = time()
             os.system(
                 f"python3 {pwd}/algorithms/{self.algorithm}/algorithm.py '{json.dumps(args)}' 1>/dev/null 2>/dev/null"
             )
+            duration = time() - start_time
+
+            with open(f"{path}/time.csv", "a", encoding="utf-8") as f:
+                f.write(f"{self.algorithm},{file_name},{duration}")