Complete Experiment

create_json.py

@@ -0,0 +1,64 @@
+import json
+import os
+from locale import atof
+import pandas as pd
+import numpy as np
+
+lines = open("results_pbc4.txt").readlines()
+results = []
+for line in lines:
+    (
+        dataset,
+        accuracy,
+        accuracy_std,
+        time_spent,
+        time_std,
+        nodes,
+        nodes_std,
+    ) = line.split(";")
+    data = pd.read_csv(
+        os.path.join("data.r", f"{dataset}_R.dat"),
+        sep="\t",
+        index_col=0,
+    )
+    X = data.drop("clase", axis=1).to_numpy()
+    y = data["clase"].to_numpy()
+    samples, features = X.shape
+    n_classes = len(np.unique(y))
+    record = {}
+    record["dataset"] = dataset
+    record["samples"] = samples
+    record["features"] = features
+    record["classes"] = n_classes
+    record["hyperparameters"] = {}
+    record["nodes"] = atof(nodes)
+    record["leaves"] = 0.0
+    record["depth"] = 0.0
+    record["score"] = atof(accuracy)
+    record["score_std"] = atof(accuracy_std)
+    record["time"] = atof(time_spent)
+    record["time_std"] = atof(time_std)
+    results.append(record)
+
+output = {}
+output["score_name"] = "accuracy"
+output["title"] = "PBC4cip Ensemble execution"
+output["model"] = "pbc4cip"
+output["version"] = "1.0"
+output["stratified"] = True
+output["folds"] = 5
+output["date"] = "2024-12-23"
+output["time"] = "17:10:00"
+output["duration"] = 1.25 * 3600
+output["seeds"] = [57, 31, 1714, 17, 23, 79, 83, 97, 7, 1]
+output["platform"] = "HugeFedora"
+output["language"] = "java"
+output["language_version"] = "openjdk 17.0.1"
+output["discretization_algorithm"] = "mdlp"
+output["smooth_strategy"] = "ORIGINAL"
+output["discretized"] = False
+output["results"] = results
+file_name = "results_accuracy_pbc4cip_HugeFedora_2024-12-23_17:10:00_1.json"
+with open(file_name, "w") as f:
+    json.dump(output, f)
+print(f"File {file_name} created.")