#!/usr/bin/env python3 import argparse import csv import json import math from pathlib import Path def parse_args(): parser = argparse.ArgumentParser(description="Plot benchmark metrics from benchmark_history.csv") base_dir = Path(__file__).resolve().parent parser.add_argument( "--figure", choices=["panel", "summary"], default="panel", help="Figure type: panel (paper-style multi-panel) or summary (seed robustness only).", ) parser.add_argument( "--history", default=str(base_dir / "results" / "benchmark_history.csv"), help="Path to benchmark_history.csv", ) parser.add_argument( "--ks-per-feature", default=str(base_dir / "results" / "ks_per_feature.csv"), help="Path to ks_per_feature.csv", ) parser.add_argument( "--data-shift", default=str(base_dir / "results" / "data_shift_stats.csv"), help="Path to data_shift_stats.csv", ) parser.add_argument( "--metrics-history", default=str(base_dir / "results" / "metrics_history.csv"), help="Path to metrics_history.csv", ) parser.add_argument( "--filtered-metrics", default=str(base_dir / "results" / "filtered_metrics.json"), help="Path to filtered_metrics.json (optional).", ) parser.add_argument( "--out", default="", help="Output SVG path (default depends on --figure).", ) parser.add_argument( "--engine", choices=["auto", "matplotlib", "svg"], default="auto", help="Plotting engine: auto prefers matplotlib if available; otherwise uses pure-SVG.", ) return parser.parse_args() def mean_std(vals): m = sum(vals) / len(vals) if len(vals) == 1: return m, 0.0 v = sum((x - m) * (x - m) for x in vals) / (len(vals) - 1) return m, math.sqrt(v) def svg_escape(s): return ( str(s) .replace("&", "&") .replace("<", "<") .replace(">", ">") .replace('"', """) .replace("'", "'") ) def clamp(v, lo, hi): if v < lo: return lo if v > hi: return hi return v def lerp(a, b, t): return a + (b - a) * t def hex_to_rgb(h): h = h.lstrip("#") return int(h[0:2], 16), int(h[2:4], 16), int(h[4:6], 16) def rgb_to_hex(r, g, b): return "#{:02x}{:02x}{:02x}".format(int(clamp(r, 0, 255)), int(clamp(g, 0, 255)), int(clamp(b, 0, 255))) def diverging_color(v, vmin=-2.0, vmax=2.0, cold="#2563eb", hot="#ef4444", mid="#ffffff"): v = clamp(v, vmin, vmax) if v >= 0: t = 0.0 if vmax == 0 else v / vmax r0, g0, b0 = hex_to_rgb(mid) r1, g1, b1 = hex_to_rgb(hot) return rgb_to_hex(lerp(r0, r1, t), lerp(g0, g1, t), lerp(b0, b1, t)) t = 0.0 if vmin == 0 else (-v) / (-vmin) r0, g0, b0 = hex_to_rgb(mid) r1, g1, b1 = hex_to_rgb(cold) return rgb_to_hex(lerp(r0, r1, t), lerp(g0, g1, t), lerp(b0, b1, t)) def plot_matplotlib(rows, seeds, metrics, out_path): import matplotlib.pyplot as plt try: plt.style.use("seaborn-v0_8-whitegrid") except Exception: pass fig, axes = plt.subplots(nrows=len(metrics), ncols=1, figsize=(8.6, 4.6), sharex=False) if len(metrics) == 1: axes = [axes] point_color = "#3b82f6" band_color = "#ef4444" grid_color = "#e5e7eb" axis_color = "#111827" for ax, (key, title) in zip(axes, metrics, strict=True): vals = [r[key] for r in rows] m, s = mean_std(vals) vmin = min(vals + [m - s]) vmax = max(vals + [m + s]) if vmax == vmin: vmax = vmin + 1.0 vr = vmax - vmin vmin -= 0.20 * vr vmax += 0.20 * vr y0 = 0.0 jitter = [-0.08, 0.0, 0.08] ys = [(y0 + jitter[i % len(jitter)]) for i in range(len(vals))] ax.axvspan(m - s, m + s, color=band_color, alpha=0.10, linewidth=0) ax.axvline(m, color=band_color, linewidth=2.2) ax.scatter(vals, ys, s=46, color=point_color, edgecolors="white", linewidths=1.0, zorder=3) for x, y, seed in zip(vals, ys, seeds, strict=True): ax.annotate( str(seed), (x, y), textcoords="offset points", xytext=(0, 10), ha="center", va="bottom", fontsize=8, color=axis_color, ) ax.set_title(title, loc="left", fontsize=11, color=axis_color, pad=8) ax.set_yticks([]) ax.set_ylim(-0.35, 0.35) ax.set_xlim(vmin, vmax) ax.grid(True, axis="x", color=grid_color) ax.grid(False, axis="y") ax.text( 0.99, 0.80, "mean={m:.4f} ± {s:.4f}".format(m=m, s=s), transform=ax.transAxes, ha="right", va="center", fontsize=9, color="#374151", ) fig.suptitle("Benchmark Metrics (3 seeds) · lower is better", fontsize=12, color=axis_color, y=0.98) fig.tight_layout(rect=(0, 0, 1, 0.95)) out_path.parent.mkdir(parents=True, exist_ok=True) fig.savefig(out_path, format="svg") plt.close(fig) def plot_svg(rows, seeds, metrics, out_path): W, H = 980, 440 pad_l, pad_r, pad_t, pad_b = 200, 30, 74, 36 row_gap = 52 row_h = (H - pad_t - pad_b - row_gap * (len(metrics) - 1)) / len(metrics) bg = "#ffffff" axis = "#2b2b2b" grid = "#e9e9e9" band = "#d62728" band_fill = "#d62728" point = "#1f77b4" text = "#111111" subtle = "#666666" parts = [] parts.append( "".format( w=W, h=H ) ) parts.append("".format(w=W, h=H, bg=bg)) parts.append( "Benchmark Metrics (3 seeds)".format( x=W / 2, c=text ) ) parts.append( "line: mean · band: ±1 std · dots: runs · lower is better".format( x=W / 2, c=subtle ) ) parts.append( "seeds: {s}".format( x=W / 2, c=subtle, s=svg_escape(", ".join(seeds)) ) ) plot_x0 = pad_l plot_x1 = W - pad_r for mi, (key, title) in enumerate(metrics): y0 = pad_t + mi * (row_h + row_gap) y1 = y0 + row_h yc = (y0 + y1) / 2 vals = [r[key] for r in rows] m, s = mean_std(vals) vmin = min(vals + [m - s]) vmax = max(vals + [m + s]) if vmax == vmin: vmax = vmin + 1.0 vr = vmax - vmin vmin -= 0.20 * vr vmax += 0.20 * vr def X(v): return plot_x0 + (v - vmin) * (plot_x1 - plot_x0) / (vmax - vmin) if mi % 2 == 1: parts.append( "".format( x=0, y=y0 - 8, w=W, h=row_h + 16 ) ) parts.append( "{t}".format( x=pad_l - 12, y=yc + 4, c=text, t=svg_escape(title) ) ) for k in range(6): xx = plot_x0 + k * (plot_x1 - plot_x0) / 5 parts.append( "".format( x=xx, y0=y0 + 4, y1=y1 - 4, c=grid ) ) val = vmin + k * (vmax - vmin) / 5 parts.append( "{v:.4f}".format( x=xx, y=y1 + 28, c=subtle, v=val ) ) parts.append( "".format( x0=plot_x0, x1=plot_x1, y=yc, c=axis ) ) x_lo = X(m - s) x_hi = X(m + s) parts.append( "".format( x=min(x_lo, x_hi), y=yc - 14, w=abs(x_hi - x_lo), h=28, c=band_fill ) ) xm = X(m) parts.append( "".format( x=xm, y0=yc - 16, y1=yc + 16, c=band ) ) parts.append( "mean={m:.4f}±{s:.4f}".format( x=xm + 6, y=yc - 18, c=subtle, m=m, s=s ) ) for i, r in enumerate(rows): jitter = ((i * 37) % 11 - 5) * 0.9 xx = X(r[key]) yy = yc + jitter parts.append("".format(x=xx, y=yy, c=point)) parts.append("") out_path.parent.mkdir(parents=True, exist_ok=True) out_path.write_text("\n".join(parts), encoding="utf-8") def read_csv_rows(path): p = Path(path) if not p.exists(): return [] with p.open("r", encoding="utf-8", newline="") as f: reader = csv.DictReader(f) return list(reader) def read_json(path): p = Path(path) if not p.exists(): return None return json.loads(p.read_text(encoding="utf-8")) def parse_float(s): if s is None: return None ss = str(s).strip() if ss == "" or ss.lower() == "none" or ss.lower() == "nan": return None return float(ss) def zscores(vals): if not vals: return [] m = sum(vals) / len(vals) v = sum((x - m) * (x - m) for x in vals) / len(vals) s = math.sqrt(v) if s == 0: return [0.0 for _ in vals] return [(x - m) / s for x in vals] def panel_svg(bh_rows, ks_rows, shift_rows, hist_rows, filtered_metrics, out_path): W, H = 1400, 900 margin = 42 gap = 26 panel_w = (W - margin * 2 - gap) / 2 panel_h = (H - margin * 2 - gap) / 2 bg = "#ffffff" ink = "#111827" subtle = "#6b7280" border = "#e5e7eb" grid = "#eef2f7" blue = "#3b82f6" red = "#ef4444" green = "#10b981" def panel_rect(x, y): return "".format( x=x, y=y, w=panel_w, h=panel_h, b=border ) def text(x, y, s, size=12, anchor="start", color=ink, weight="normal"): return "{t}".format( x=x, y=y, a=anchor, fs=size, c=color, w=weight, t=svg_escape(s) ) def line(x1, y1, x2, y2, color=border, width=1.0, dash=None, opacity=1.0, cap="round"): extra = "" if dash: extra += " stroke-dasharray='{d}'".format(d=dash) if opacity != 1.0: extra += " stroke-opacity='{o}'".format(o=opacity) return "".format( x1=x1, y1=y1, x2=x2, y2=y2, c=color, w=width, cap=cap, extra=extra ) def round_box(x, y, w, h, fill="#ffffff", stroke=border, sw=1.2, rx=12): return "".format( x=x, y=y, w=w, h=h, rx=rx, f=fill, s=stroke, sw=sw ) def arrow(x1, y1, x2, y2, color=ink, width=1.8): ang = math.atan2(y2 - y1, x2 - x1) ah = 10.0 aw = 5.0 hx = x2 - ah * math.cos(ang) hy = y2 - ah * math.sin(ang) px = aw * math.sin(ang) py = -aw * math.cos(ang) p1x, p1y = hx + px, hy + py p2x, p2y = hx - px, hy - py return ( "" "" ).format(x1=x1, y1=y1, x2=x2, y2=y2, c=color, w=width, p1x=p1x, p1y=p1y, p2x=p2x, p2y=p2y) parts = [] parts.append( "".format(w=W, h=H) ) parts.append("".format(w=W, h=H, bg=bg)) parts.append(text(W / 2, 32, "Benchmark Overview (HAI Security Dataset)", size=18, anchor="middle", weight="bold")) parts.append( text( W / 2, 54, "A: workflow · B: per-feature KS · C: train-file mean shift · D: seed robustness and metric history", size=11, anchor="middle", color=subtle, ) ) xA, yA = margin, margin + 36 xB, yB = margin + panel_w + gap, margin + 36 xC, yC = margin, margin + 36 + panel_h + gap xD, yD = margin + panel_w + gap, margin + 36 + panel_h + gap parts.append(panel_rect(xA, yA)) parts.append(panel_rect(xB, yB)) parts.append(panel_rect(xC, yC)) parts.append(panel_rect(xD, yD)) def panel_label(x, y, letter, title): parts.append(text(x + 18, y + 28, letter, size=16, weight="bold")) parts.append(text(x + 44, y + 28, title, size=14, weight="bold")) panel_label(xA, yA, "A", "Typed Hybrid Generation") panel_label(xB, yB, "B", "Feature-Level Distribution Fidelity") panel_label(xC, yC, "C", "Dataset Shift Across Training Files") panel_label(xD, yD, "D", "Robustness Across Seeds") ax0 = xA + 22 ay0 = yA + 56 aw0 = panel_w - 44 ah0 = panel_h - 78 box_h = 56 box_w = (aw0 - 3 * 26) / 4 by = ay0 + (ah0 - box_h) / 2 - 14 boxes = [ ("HAI windows\n(L=96)", "#f8fafc"), ("Typed\ndecomposition", "#f8fafc"), ("Hybrid\ngenerator", "#f8fafc"), ("Synthetic\nwindows", "#f8fafc"), ] for i, (lbl, fill) in enumerate(boxes): bx = ax0 + i * (box_w + 26) parts.append(round_box(bx, by, box_w, box_h, fill=fill)) for j, line_txt in enumerate(lbl.split("\n")): parts.append(text(bx + box_w / 2, by + 22 + j * 16, line_txt, size=11, anchor="middle", weight="bold" if j == 0 else "normal")) if i < len(boxes) - 1: parts.append(arrow(bx + box_w, by + box_h / 2, bx + box_w + 26, by + box_h / 2, color=subtle, width=1.6)) hx = ax0 + 2 * (box_w + 26) hy = by + box_h + 18 parts.append(text(hx + 6, hy - 6, "Type-aware routes", size=10, color=subtle, weight="bold")) inner_w = box_w inner_gap = 10 inner_h = 32 inner_y = hy inner_colors = [("#e0f2fe", blue, "Trend (det.)"), ("#fee2e2", red, "Residual (DDPM)"), ("#dcfce7", green, "Discrete head")] for k, (fill, stroke, name) in enumerate(inner_colors): iy = inner_y + k * (inner_h + inner_gap) parts.append(round_box(hx, iy, inner_w, inner_h, fill=fill, stroke=stroke, sw=1.4, rx=10)) parts.append(text(hx + 10, iy + 20, name, size=10, color=ink, weight="bold")) parts.append(arrow(hx + inner_w / 2, by + box_h, hx + inner_w / 2, inner_y, color=subtle, width=1.4)) parts.append( text( xA + 22, yA + panel_h - 18, "Separation aligns metrics with data types: KS (continuous), JSD (discrete), lag-1 (temporal).", size=10, color=subtle, ) ) bx0 = xB + 22 by0 = yB + 62 bw0 = panel_w - 44 bh0 = panel_h - 86 ks_sorted = sorted( [ { "feature": r.get("feature", ""), "ks": parse_float(r.get("ks")), "gen_frac_at_min": parse_float(r.get("gen_frac_at_min")), "gen_frac_at_max": parse_float(r.get("gen_frac_at_max")), } for r in ks_rows if parse_float(r.get("ks")) is not None ], key=lambda x: x["ks"], reverse=True, ) top_n = 14 if len(ks_sorted) >= 14 else len(ks_sorted) top = ks_sorted[:top_n] dropped = [] if isinstance(filtered_metrics, dict): for d in filtered_metrics.get("dropped_features", []) or []: feat = d.get("feature") if feat: dropped.append(feat) parts.append(text(bx0, by0 - 8, "Top-{n} KS outliers (lower is better)".format(n=top_n), size=11, color=subtle)) if dropped: parts.append(text(bx0 + bw0, by0 - 8, "dropped: {d}".format(d=", ".join(dropped)), size=10, anchor="end", color=subtle)) chart_y0 = by0 + 16 chart_h = bh0 - 32 label_w = 180 x0 = bx0 + label_w x1 = bx0 + bw0 - 16 row_h = chart_h / max(1, top_n) for t in range(6): xx = x0 + (x1 - x0) * (t / 5) parts.append(line(xx, chart_y0, xx, chart_y0 + chart_h, color=grid, width=1.0)) parts.append(text(xx, chart_y0 + chart_h + 18, "{:.1f}".format(t / 5), size=9, anchor="middle", color=subtle)) parts.append(line(x0, chart_y0 + chart_h, x1, chart_y0 + chart_h, color=border, width=1.2, cap="butt")) for i, r in enumerate(top): fy = chart_y0 + i * row_h + row_h / 2 feat = r["feature"] ks = r["ks"] gmin = r["gen_frac_at_min"] or 0.0 gmax = r["gen_frac_at_max"] or 0.0 collapsed = (gmin >= 0.98) or (gmax >= 0.98) bar_color = "#0ea5e9" if not collapsed else "#fb7185" parts.append(text(x0 - 10, fy + 4, feat, size=9, anchor="end", color=ink)) parts.append(text(x1 + 8, fy + 4, "{:.3f}".format(ks), size=9, anchor="start", color=subtle)) w = (x1 - x0) * clamp(ks, 0.0, 1.0) parts.append( "".format( x=x0, y=fy - row_h * 0.34, w=w, h=row_h * 0.68, c=bar_color ) ) if collapsed: parts.append( text( x0 + min(w, (x1 - x0) - 10), fy + 4, "collapse", size=8, anchor="end", color="#7f1d1d", weight="bold", ) ) cx0 = xC + 22 cy0 = yC + 64 cw0 = panel_w - 44 ch0 = panel_h - 88 if shift_rows: cols = list(shift_rows[0].keys()) else: cols = [] mean_cols = [c for c in cols if c.startswith("mean_")] wanted = ["mean_P1_FT01", "mean_P1_LIT01", "mean_P1_PIT01", "mean_P2_CO_rpm", "mean_P3_LIT01", "mean_P4_ST_PT01"] feats = [c for c in wanted if c in mean_cols] files = [r.get("file", "") for r in shift_rows] sample_rows = [parse_float(r.get("sample_rows")) for r in shift_rows] feat_vals = {c: [parse_float(r.get(c)) or 0.0 for r in shift_rows] for c in feats} feat_z = {c: zscores(vs) for c, vs in feat_vals.items()} parts.append(text(cx0, cy0 - 10, "Mean shift (z-score) across train files", size=11, color=subtle)) if files: parts.append(text(cx0 + cw0, cy0 - 10, "rows: {r}".format(r=", ".join(str(int(x)) for x in sample_rows if x is not None)), size=10, anchor="end", color=subtle)) heat_x0 = cx0 + 160 heat_y0 = cy0 + 10 heat_w = cw0 - 180 heat_h = ch0 - 44 n_rows = max(1, len(files)) n_cols = max(1, len(feats)) cell_w = heat_w / n_cols cell_h = heat_h / n_rows for j, c in enumerate(feats): label = c.replace("mean_", "") parts.append(text(heat_x0 + j * cell_w + cell_w / 2, heat_y0 - 8, label, size=9, anchor="middle", color=ink, weight="bold")) for i, f in enumerate(files): fy = heat_y0 + i * cell_h + cell_h / 2 parts.append(text(cx0 + 140, fy + 4, f, size=9, anchor="end", color=ink)) for i in range(n_rows + 1): yy = heat_y0 + i * cell_h parts.append(line(heat_x0, yy, heat_x0 + heat_w, yy, color=border, width=1.0, cap="butt")) for j in range(n_cols + 1): xx = heat_x0 + j * cell_w parts.append(line(xx, heat_y0, xx, heat_y0 + heat_h, color=border, width=1.0, cap="butt")) for i in range(len(files)): for j, c in enumerate(feats): z = feat_z[c][i] if i < len(feat_z[c]) else 0.0 fill = diverging_color(z, vmin=-2.0, vmax=2.0) parts.append( "".format( x=heat_x0 + j * cell_w + 0.6, y=heat_y0 + i * cell_h + 0.6, w=cell_w - 1.2, h=cell_h - 1.2, c=fill ) ) parts.append(text(heat_x0 + j * cell_w + cell_w / 2, heat_y0 + i * cell_h + cell_h / 2 + 4, "{:+.2f}".format(z), size=9, anchor="middle", color=ink)) lx0 = heat_x0 ly0 = heat_y0 + heat_h + 18 parts.append(text(cx0 + 140, ly0 + 4, "z", size=9, anchor="end", color=subtle)) grad_w = 240 for k in range(25): t = k / 24 z = lerp(-2.0, 2.0, t) fill = diverging_color(z) parts.append( "".format( x=lx0 + k * (grad_w / 25), y=ly0 - 6, w=(grad_w / 25) + 0.2, c=fill ) ) parts.append(text(lx0, ly0 + 16, "-2", size=9, color=subtle)) parts.append(text(lx0 + grad_w / 2, ly0 + 16, "0", size=9, anchor="middle", color=subtle)) parts.append(text(lx0 + grad_w, ly0 + 16, "+2", size=9, anchor="end", color=subtle)) dx0 = xD + 22 dy0 = yD + 62 dw0 = panel_w - 44 dh0 = panel_h - 86 metrics = [ ("avg_ks", "KS (cont.)"), ("avg_jsd", "JSD (disc.)"), ("avg_lag1_diff", "Abs Δ lag-1"), ] bh_rows = sorted(bh_rows, key=lambda r: r.get("seed", 0)) seeds = [str(r.get("seed", "")) for r in bh_rows] parts.append(text(dx0, dy0 - 10, "Seed robustness (mean ± 1 std; dots: seeds)", size=11, color=subtle)) if seeds: parts.append(text(dx0 + dw0, dy0 - 10, "seeds: {s}".format(s=", ".join(seeds)), size=10, anchor="end", color=subtle)) spark_h = 48 spark_gap = 10 forest_y0 = dy0 + spark_h * 3 + spark_gap * 2 + 18 forest_h = dh0 - (spark_h * 3 + spark_gap * 2 + 28) hist_clean = [] for r in hist_rows: ks = parse_float(r.get("avg_ks")) jsd = parse_float(r.get("avg_jsd")) lag = parse_float(r.get("avg_lag1_diff")) if ks is None or jsd is None or lag is None: continue hist_clean.append({"avg_ks": ks, "avg_jsd": jsd, "avg_lag1_diff": lag}) if hist_clean: for mi, (k, title) in enumerate(metrics): y0 = dy0 + mi * (spark_h + spark_gap) y1 = y0 + spark_h parts.append(round_box(dx0, y0, dw0, spark_h, fill="#f9fafb", stroke=border, sw=1.0, rx=10)) parts.append(text(dx0 + 10, y0 + 18, title + " history", size=10, color=subtle, weight="bold")) vals = [r[k] for r in hist_clean] vmin = min(vals) vmax = max(vals) if vmax == vmin: vmax = vmin + 1.0 px0 = dx0 + 130 px1 = dx0 + dw0 - 12 py0 = y0 + 30 py1 = y1 - 10 parts.append(line(px0, py1, px1, py1, color=border, width=1.0, cap="butt")) parts.append(line(px0, py0, px0, py1, color=border, width=1.0, cap="butt")) pts = [] n = len(vals) for i, v in enumerate(vals): x = px0 + (px1 - px0) * (i / max(1, n - 1)) y = py1 - (v - vmin) * (py1 - py0) / (vmax - vmin) pts.append((x, y)) d = "M " + " L ".join("{:.1f} {:.1f}".format(x, y) for x, y in pts) parts.append("".format(d=d, c=blue if mi == 0 else (red if mi == 1 else green))) parts.append(text(dx0 + dw0 - 12, y0 + 18, "{:.3f}".format(vals[-1]), size=10, anchor="end", color=ink, weight="bold")) fx0 = dx0 fy0 = forest_y0 fw0 = dw0 fh0 = forest_h x0 = fx0 + 190 x1 = fx0 + fw0 - 18 for t in range(6): xx = x0 + (x1 - x0) * (t / 5) parts.append(line(xx, fy0, xx, fy0 + fh0, color=grid, width=1.0)) for mi, (key, title) in enumerate(metrics): y0 = fy0 + mi * (fh0 / 3) y1 = fy0 + (mi + 1) * (fh0 / 3) yc = (y0 + y1) / 2 vals = [r.get(key) for r in bh_rows if r.get(key) is not None] if not vals: continue m, s = mean_std(vals) vmin = min(vals + [m - s]) vmax = max(vals + [m + s]) if vmax == vmin: vmax = vmin + 1.0 vr = vmax - vmin vmin -= 0.15 * vr vmax += 0.15 * vr def X(v): return x0 + (v - vmin) * (x1 - x0) / (vmax - vmin) parts.append(text(x0 - 14, yc + 4, title, size=11, anchor="end", color=ink, weight="bold")) parts.append(line(x0, yc, x1, yc, color=border, width=1.2, cap="butt")) parts.append( "".format( x=X(m - s), y=yc - 10, w=max(1.0, X(m + s) - X(m - s)), c=red ) ) parts.append(line(X(m), yc - 14, X(m), yc + 14, color=red, width=2.4)) parts.append(text(x1, yc - 16, "mean={m:.4f}±{s:.4f}".format(m=m, s=s), size=9, anchor="end", color=subtle)) for i, v in enumerate(vals): jitter = ((i * 37) % 9 - 4) * 1.2 parts.append("".format(x=X(v), y=yc + jitter, c=blue)) parts.append("") out_path.parent.mkdir(parents=True, exist_ok=True) out_path.write_text("\n".join(parts), encoding="utf-8") def panel_matplotlib(bh_rows, ks_rows, shift_rows, hist_rows, filtered_metrics, out_path): import matplotlib.pyplot as plt import matplotlib.patches as patches try: plt.style.use("seaborn-v0_8-whitegrid") except Exception: pass fig = plt.figure(figsize=(13.6, 8.6)) gs = fig.add_gridspec(2, 2, wspace=0.18, hspace=0.22) axA = fig.add_subplot(gs[0, 0]) axB = fig.add_subplot(gs[0, 1]) axC = fig.add_subplot(gs[1, 0]) axD = fig.add_subplot(gs[1, 1]) fig.suptitle("Benchmark Overview (HAI Security Dataset)", fontsize=16, y=0.98) axA.set_title("A Typed Hybrid Generation", loc="left", fontsize=12, fontweight="bold") axA.axis("off") axA.set_xlim(0, 1) axA.set_ylim(0, 1) box_y = 0.55 box_w = 0.18 box_h = 0.16 x_positions = [0.06, 0.30, 0.54, 0.78] labels = ["HAI windows\n(L=96)", "Typed\ndecomposition", "Hybrid\ngenerator", "Synthetic\nwindows"] for x, lbl in zip(x_positions, labels, strict=True): axA.add_patch(patches.FancyBboxPatch((x, box_y), box_w, box_h, boxstyle="round,pad=0.02,rounding_size=0.02", facecolor="#f8fafc", edgecolor="#e5e7eb")) axA.text(x + box_w / 2, box_y + box_h / 2, lbl, ha="center", va="center", fontsize=10, fontweight="bold") for i in range(3): x1 = x_positions[i] + box_w x2 = x_positions[i + 1] axA.annotate("", xy=(x2, box_y + box_h / 2), xytext=(x1, box_y + box_h / 2), arrowprops=dict(arrowstyle="-|>", lw=1.4, color="#6b7280")) hx = x_positions[2] hy = 0.20 axA.text(hx, hy + 0.27, "Type-aware routes", fontsize=9, color="#6b7280", fontweight="bold") inner = [("Trend (det.)", "#e0f2fe", "#3b82f6"), ("Residual (DDPM)", "#fee2e2", "#ef4444"), ("Discrete head", "#dcfce7", "#10b981")] for k, (name, fc, ec) in enumerate(inner): y = hy + 0.18 - k * 0.11 axA.add_patch(patches.FancyBboxPatch((hx, y), box_w, 0.08, boxstyle="round,pad=0.02,rounding_size=0.02", facecolor=fc, edgecolor=ec, lw=1.2)) axA.text(hx + 0.01, y + 0.04, name, ha="left", va="center", fontsize=9, fontweight="bold") axA.annotate("", xy=(hx + box_w / 2, hy + 0.20), xytext=(hx + box_w / 2, box_y), arrowprops=dict(arrowstyle="-|>", lw=1.2, color="#6b7280")) axA.text(0.06, 0.06, "Metrics align with types: KS (continuous), JSD (discrete), lag-1 (temporal).", fontsize=9, color="#6b7280") axB.set_title("B Feature-Level Distribution Fidelity", loc="left", fontsize=12, fontweight="bold") ks_sorted = sorted( [ { "feature": r.get("feature", ""), "ks": parse_float(r.get("ks")), "gen_frac_at_min": parse_float(r.get("gen_frac_at_min")) or 0.0, "gen_frac_at_max": parse_float(r.get("gen_frac_at_max")) or 0.0, } for r in ks_rows if parse_float(r.get("ks")) is not None ], key=lambda x: x["ks"], reverse=True, ) top = ks_sorted[:14] feats = [r["feature"] for r in top][::-1] vals = [r["ks"] for r in top][::-1] collapsed = [((r["gen_frac_at_min"] >= 0.98) or (r["gen_frac_at_max"] >= 0.98)) for r in top][::-1] colors = ["#fb7185" if c else "#0ea5e9" for c in collapsed] axB.barh(feats, vals, color=colors) axB.set_xlabel("KS (lower is better)") axB.set_xlim(0, 1.0) if isinstance(filtered_metrics, dict) and filtered_metrics.get("dropped_features"): dropped = ", ".join(d.get("feature", "") for d in filtered_metrics["dropped_features"] if d.get("feature")) if dropped: axB.text(0.99, 0.02, "dropped: {d}".format(d=dropped), transform=axB.transAxes, ha="right", va="bottom", fontsize=9, color="#6b7280") axC.set_title("C Dataset Shift Across Training Files", loc="left", fontsize=12, fontweight="bold") if shift_rows: cols = list(shift_rows[0].keys()) else: cols = [] mean_cols = [c for c in cols if c.startswith("mean_")] wanted = ["mean_P1_FT01", "mean_P1_LIT01", "mean_P1_PIT01", "mean_P2_CO_rpm", "mean_P3_LIT01", "mean_P4_ST_PT01"] feats = [c for c in wanted if c in mean_cols] files = [r.get("file", "") for r in shift_rows] M = [] for c in feats: M.append([parse_float(r.get(c)) or 0.0 for r in shift_rows]) if M and files and feats: Z = list(zip(*[zscores(col) for col in M], strict=True)) im = axC.imshow(Z, aspect="auto", cmap="coolwarm", vmin=-2, vmax=2) axC.set_yticks(range(len(files))) axC.set_yticklabels(files) axC.set_xticks(range(len(feats))) axC.set_xticklabels([f.replace("mean_", "") for f in feats], rotation=25, ha="right") axC.set_ylabel("Train file") axC.set_xlabel("Feature mean z-score") fig.colorbar(im, ax=axC, fraction=0.046, pad=0.04) else: axC.axis("off") axC.text(0.5, 0.5, "missing data_shift_stats.csv", ha="center", va="center", fontsize=11, color="#6b7280") axD.set_title("D Robustness Across Seeds", loc="left", fontsize=12, fontweight="bold") axD.axis("off") axD.set_xlim(0, 1) axD.set_ylim(0, 1) metrics = [("avg_ks", "KS (cont.)"), ("avg_jsd", "JSD (disc.)"), ("avg_lag1_diff", "Abs Δ lag-1")] bh_rows = sorted(bh_rows, key=lambda r: r.get("seed", 0)) for mi, (k, title) in enumerate(metrics): vals = [r.get(k) for r in bh_rows if r.get(k) is not None] if not vals: continue m, s = mean_std(vals) y = 0.78 - mi * 0.22 axD.text(0.04, y, title, fontsize=10, fontweight="bold", va="center") x0 = 0.42 x1 = 0.96 vmin = min(vals + [m - s]) vmax = max(vals + [m + s]) if vmax == vmin: vmax = vmin + 1.0 vr = vmax - vmin vmin -= 0.15 * vr vmax += 0.15 * vr def X(v): return x0 + (v - vmin) * (x1 - x0) / (vmax - vmin) axD.add_patch(patches.FancyBboxPatch((X(m - s), y - 0.03), max(0.002, X(m + s) - X(m - s)), 0.06, boxstyle="round,pad=0.01,rounding_size=0.02", facecolor="#ef4444", alpha=0.12, edgecolor="none")) axD.plot([X(m), X(m)], [y - 0.05, y + 0.05], color="#ef4444", lw=2.2) jit = [-0.03, 0.0, 0.03] for i, v in enumerate(vals): axD.scatter([X(v)], [y + jit[i % len(jit)]], s=40, color="#3b82f6", edgecolor="white", linewidth=0.9, zorder=3) axD.text(0.96, y + 0.07, "mean={m:.4f}±{s:.4f}".format(m=m, s=s), fontsize=9, color="#6b7280", ha="right") fig.tight_layout(rect=(0, 0, 1, 0.96)) out_path.parent.mkdir(parents=True, exist_ok=True) fig.savefig(out_path, format="svg") plt.close(fig) def main(): args = parse_args() hist_path = Path(args.history) if not hist_path.exists(): raise SystemExit("missing history file: %s" % hist_path) rows = [] with hist_path.open("r", encoding="utf-8", newline="") as f: reader = csv.DictReader(f) for r in reader: rows.append( { "run_name": r["run_name"], "seed": int(r["seed"]), "avg_ks": float(r["avg_ks"]), "avg_jsd": float(r["avg_jsd"]), "avg_lag1_diff": float(r["avg_lag1_diff"]), } ) if not rows: raise SystemExit("no rows in history file: %s" % hist_path) rows = sorted(rows, key=lambda x: x["seed"]) seeds = [str(r["seed"]) for r in rows] metrics = [ ("avg_ks", "KS (continuous)"), ("avg_jsd", "JSD (discrete)"), ("avg_lag1_diff", "Abs Δ lag-1 autocorr"), ] if args.out: out_path = Path(args.out) else: if args.figure == "panel": out_path = Path(__file__).resolve().parent.parent / "figures" / "benchmark_panel.svg" else: out_path = Path(__file__).resolve().parent.parent / "figures" / "benchmark_metrics.svg" if args.figure == "summary": if args.engine in {"auto", "matplotlib"}: try: plot_matplotlib(rows, seeds, metrics, out_path) print("saved", out_path) return except Exception: if args.engine == "matplotlib": raise plot_svg(rows, seeds, metrics, out_path) print("saved", out_path) return ks_rows = read_csv_rows(args.ks_per_feature) shift_rows = read_csv_rows(args.data_shift) mh_rows = read_csv_rows(args.metrics_history) fm = read_json(args.filtered_metrics) bh_rows = [{"seed": r["seed"], "avg_ks": r["avg_ks"], "avg_jsd": r["avg_jsd"], "avg_lag1_diff": r["avg_lag1_diff"]} for r in rows] if args.engine in {"auto", "matplotlib"}: try: panel_matplotlib(bh_rows, ks_rows, shift_rows, mh_rows, fm, out_path) print("saved", out_path) return except Exception: if args.engine == "matplotlib": raise panel_svg(bh_rows, ks_rows, shift_rows, mh_rows, fm, out_path) print("saved", out_path) if __name__ == "__main__": main()