kriging-rs 0.4.0

Geostatistical kriging library with WASM support
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169

#!/usr/bin/env python3
"""
Plot summaries from `accuracy_speed_report` (summary.tsv).
Requires: matplotlib

  python3 scripts/plot_accuracy_speed.py --tsv target/accuracy_sweep/summary.tsv --out target/accuracy_sweep/figs
"""
from __future__ import annotations

import argparse
import csv
from collections import defaultdict
from pathlib import Path


def load_rows(path: Path) -> list[dict[str, str]]:
    with path.open(newline="") as f:
        r = csv.DictReader(f, delimiter="\t")
        return [dict(row) for row in r]


def group_by_scenario(rows: list[dict[str, str]]) -> dict[str, list[dict[str, str]]]:
    d: dict[str, list[dict[str, str]]] = defaultdict(list)
    for row in rows:
        d[row["scenario"]].append(row)
    return dict(d)


def _f(row: dict[str, str], key: str) -> float:
    return float(row[key])


def plot_mae_rmse(
    by_scen: dict[str, list[dict[str, str]]], out: Path, title: str
) -> None:
    import matplotlib

    matplotlib.use("Agg")
    import matplotlib.pyplot as plt

    scenarios = sorted(by_scen.keys())
    paths = ["homo", "hetero"]
    x0 = list(range(len(scenarios)))
    w = 0.35
    fig, (ax0, ax1) = plt.subplots(1, 2, figsize=(10, 4))
    for i, p in enumerate(paths):
        off = (i - 0.5) * w
        mae = [_f(next(r for r in by_scen[s] if r["path"] == p), "mae_prevalence") for s in scenarios]
        rmse = [
            _f(next(r for r in by_scen[s] if r["path"] == p), "rmse_prevalence")
            for s in scenarios
        ]
        xp = [xi + off for xi in x0]
        ax0.bar(xp, mae, w, label=p)
        ax1.bar(xp, rmse, w, label=p)
    for ax, ylab in ((ax0, "MAE (prevalence)"), (ax1, "RMSE (prevalence)")):
        ax.set_xticks(x0, scenarios, rotation=15, ha="right")
        ax.set_ylabel(ylab)
        ax.grid(axis="y", alpha=0.3)
    ax0.legend()
    ax1.legend()
    fig.suptitle(title)
    fig.tight_layout()
    dest = out / "accuracy_mae_rmse.png"
    fig.savefig(dest, dpi=150)
    plt.close(fig)
    print(f"Wrote {dest}")


def plot_timings(
    by_scen: dict[str, list[dict[str, str]]], out: Path, title: str
) -> None:
    import matplotlib

    matplotlib.use("Agg")
    import matplotlib.pyplot as plt

    scenarios = sorted(by_scen.keys())
    paths = ["homo", "hetero"]
    # One grouped cluster per (scenario, path) with stacked build / pred parts
    labels: list[str] = []
    fit_ms: list[float] = []
    build_ms: list[float] = []
    ptest_ms: list[float] = []
    pgrid_ms: list[float] = []
    for s in scenarios:
        for p in paths:
            row = next(r for r in by_scen[s] if r["path"] == p)
            labels.append(f"{s}\n{p}")
            fit_ms.append(_f(row, "fit_variogram_ms"))
            build_ms.append(_f(row, "build_model_ms"))
            ptest_ms.append(_f(row, "pred_test_ms"))
            pgrid_ms.append(_f(row, "pred_grid_ms"))

    x = list(range(len(labels)))
    w = 0.55
    fig, ax = plt.subplots(figsize=(max(8, 0.45 * len(labels)), 4.5))
    ax.bar(x, fit_ms, w, label="fit variogram")
    ax.bar(x, build_ms, w, bottom=fit_ms, label="build model")
    ax.bar(
        x,
        ptest_ms,
        w,
        bottom=[a + b for a, b in zip(fit_ms, build_ms)],
        label="pred test set",
    )
    bot = [a + b + c for a, b, c in zip(fit_ms, build_ms, ptest_ms)]
    ax.bar(x, pgrid_ms, w, bottom=bot, label="pred grid")
    ax.set_xticks(x, labels, rotation=20, ha="right", fontsize=8)
    ax.set_ylabel("Time (ms)")
    ax.set_title(title)
    ax.legend(loc="upper left", fontsize=8)
    ax.grid(axis="y", alpha=0.3)
    fig.tight_layout()
    dest = out / "accuracy_timings_stacked.png"
    fig.savefig(dest, dpi=150)
    plt.close(fig)
    print(f"Wrote {dest}")


def plot_scatter_accuracy_vs_speed(
    rows: list[dict[str, str]], out: Path, title: str
) -> None:
    import matplotlib

    matplotlib.use("Agg")
    import matplotlib.pyplot as plt

    fig, ax = plt.subplots(figsize=(6, 5))
    for row in rows:
        t = _f(row, "build_model_ms") + _f(row, "pred_grid_ms")
        mae = _f(row, "mae_prevalence")
        ax.scatter(t, mae, s=60, label=f"{row['scenario']}/{row['path']}", alpha=0.85)
    ax.set_xlabel("build + pred_grid (ms)")
    ax.set_ylabel("MAE (prevalence)")
    ax.set_title(title)
    ax.grid(alpha=0.3)
    h, l = ax.get_legend_handles_labels()
    if h:
        ax.legend(fontsize=7, loc="best")
    fig.tight_layout()
    dest = out / "accuracy_vs_speed_mae.png"
    fig.savefig(dest, dpi=150)
    plt.close(fig)
    print(f"Wrote {dest}")


def main() -> None:
    ap = argparse.ArgumentParser()
    ap.add_argument("--tsv", type=Path, required=True, help="summary.tsv from accuracy_speed_report")
    ap.add_argument(
        "--out", type=Path, required=True, help="output directory for PNGs"
    )
    ap.add_argument(
        "--title", default="Binomial kriging (homo vs hetero)", help="plot title"
    )
    args = ap.parse_args()
    rows = load_rows(args.tsv)
    if not rows:
        raise SystemExit("empty TSV")
    args.out.mkdir(parents=True, exist_ok=True)
    by = group_by_scenario(rows)
    plot_mae_rmse(by, args.out, args.title)
    plot_timings(by, args.out, args.title)
    plot_scatter_accuracy_vs_speed(rows, args.out, args.title)


if __name__ == "__main__":
    main()