Add heuristic processing

data/benchmark/heuristic_processing.py

@@ -0,0 +1,119 @@
+"""Data processing for the benchmark results."""
+
+from dataclasses import dataclass, asdict
+
+import json
+import matplotlib.pyplot as plt
+import pandas as pd
+
+
+@dataclass
+class MakespanResult:
+    """Holds the makespan values for the baseline, heuristic, and rl algorithms"""
+
+    baseline: float
+    heuristic: float
+    rl: float
+
+
+@dataclass
+class ScoreResult:
+    """Holds the M score values for the baseline, heuristic, and rl algorithms."""
+
+    baseline: float
+    heuristic: float
+    rl: float
+
+
+@dataclass
+class NoiseResult:
+    """Holds the noise values for the baseline, heuristic, and rl algorithms"""
+
+    baseline: float
+    heuristic: float
+    rl: float
+
+
+def analyze_heuristic_benchmarks(in_file: str) -> None:
+    """Visualizes the benchmark results and calculates the average improvements.
+    Args:
+        in_file (str): The file containing the benchmark results.
+
+    """
+    with open(in_file, "r", encoding="utf-8") as f:
+        data: list[dict] = json.load(f)
+    for setting in data:
+        title = str(setting["setting"])
+        benchmarks = setting["benchmarks"]
+        makespans, scores, noises = [], [], []
+        # Loop through each benchmark
+        for benchmark in benchmarks:
+            makespans.append(
+                MakespanResult(
+                    baseline=benchmark["baseline"]["makespan"],
+                    heuristic=benchmark["heuristic"]["makespan"],
+                    rl=benchmark["rl"]["makespan"],
+                )
+            )
+            scores.append(
+                ScoreResult(
+                    baseline=benchmark["baseline"]["metric"],
+                    heuristic=benchmark["heuristic"]["metric"],
+                    rl=benchmark["rl"]["metric"],
+                )
+            )
+            noises.append(
+                NoiseResult(
+                    baseline=benchmark["baseline"]["noise"],
+                    heuristic=benchmark["heuristic"]["noise"],
+                    rl=benchmark["rl"]["noise"],
+                )
+            )
+
+        _plot_boxplots(title, makespans, scores, noises)
+    # Display the resulting plot
+    plt.tight_layout()
+    plt.savefig(in_file.replace(".json", ".pdf"))
+
+
+def _plot_boxplots(
+    title: str,
+    makespans: list[MakespanResult],
+    scores: list[ScoreResult],
+    noises: list[NoiseResult],
+) -> None:
+    props = {
+        "boxprops": {"facecolor": "#98c6ea", "edgecolor": "#154060"},
+        "whiskerprops": {"color": "#154060"},
+        "medianprops": {"color": "#e37222"},
+        "capprops": {"color": "#154060"},
+        "patch_artist": True,
+        "widths": 0.25,
+        "positions": [0.5, 1, 1.5],
+    }
+    makespan_data = pd.DataFrame([asdict(result) for result in makespans])
+    score_data = pd.DataFrame([asdict(result) for result in scores])
+    noise_data = pd.DataFrame([asdict(result) for result in noises])
+
+    # Plotting boxplots
+    fig, axes = plt.subplots(3, 1, figsize=(5, 7))
+    fig.suptitle(title)
+
+    makespan_data.plot(
+        kind="box",
+        ax=axes[0],
+        **props,
+    )
+    axes[0].set_title("Makespan")
+    axes[0].set_ylabel(r"Time [$\mu s$]")
+
+    score_data.plot(kind="box", ax=axes[1], **props)
+    axes[1].set_title("Score")
+    axes[1].set_ylabel(r"$\mathbf{M}$")
+
+    noise_data.plot(kind="box", ax=axes[2], **props)
+    axes[2].set_title("Noise")
+    axes[2].set_ylabel(r"$\mathbf{F}$")
+
+    for ax in axes:
+        ax.set_xticklabels(["Baseline", "Heuristic", "RL"])