File size: 3,518 Bytes
b154e4c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
from __future__ import annotations

import hashlib
from pathlib import Path
from typing import List

import numpy as np
import pandas as pd

from .calibration import EPS


def safe_read_csv(path: Path) -> pd.DataFrame:
    if not path.exists():
        return pd.DataFrame()
    return pd.read_csv(path)


def sigmoid(x):
    return 1.0 / (1.0 + np.exp(-x))


def clip01(x):
    return np.clip(x, 0.0, 1.0)


def stable_softmax(logits: np.ndarray, tau: float) -> np.ndarray:
    z = logits / max(tau, 1e-9)
    z = z - z.max(axis=1, keepdims=True)
    ez = np.exp(z)
    return ez / np.maximum(ez.sum(axis=1, keepdims=True), EPS)


def fixed_scale(series: pd.Series, low: float, high: float, invert: bool = False) -> pd.Series:
    s = series.astype(float).fillna(low)
    out = ((s - low) / max(high - low, EPS)).clip(0.0, 1.0)
    return 1.0 - out if invert else out


def infer_family(type_name: str) -> str:
    if "Lband" in type_name:
        return "Lband"
    if "Cband" in type_name:
        return "Cband"
    if "Xband" in type_name:
        return "Xband"
    if type_name.startswith("SC_"):
        return "SC"
    return "unknown"


def infer_type_from_family(family: str) -> str:
    return {
        "Lband": "NC_pillbox_Lband",
        "Cband": "NC_Cband_TW",
        "Xband": "NC_Xband_SW",
        "SC": "SC_elliptical_1p3GHz",
    }.get(family, "NC_pillbox_Lband")


def family_label(family: str) -> str:
    return {
        "Lband": "L-band",
        "Cband": "C-band",
        "Xband": "X-band",
        "SC": "SC",
    }.get(family, str(family))


def fmt_float(x, nd=3):
    return "—" if pd.isna(x) else f"{float(x):.{nd}f}"


def fmt_sci(x, nd=2):
    return "—" if pd.isna(x) else f"{float(x):.{nd}e}"


def fmt_text(x):
    return "—" if pd.isna(x) or str(x).strip() == "" else str(x)


def split_flags(flag_str: str) -> List[str]:
    if not isinstance(flag_str, str) or not flag_str.strip():
        return []
    return [x.strip() for x in flag_str.split("|") if x.strip()]


def warning_badges(flag_str: str) -> str:
    flags = split_flags(flag_str)
    if not flags:
        return "—"
    badge_map = {
        "pulse_heating_yellow": "⚠ thermal",
        "BDR_yellow": "⚠ BDR",
    }
    return ", ".join([badge_map.get(f, f"⚠ {f}") for f in flags])


def _fmt_gene_value(x) -> str:
    if pd.isna(x):
        return "nan"
    return f"{float(x):.6f}"


def make_candidate_id(row: pd.Series) -> str:
    fields = [
        fmt_text(row.get("type", "")),
        _fmt_gene_value(row.get("freq_GHz")),
        _fmt_gene_value(row.get("R_over_Q_ohm")),
        _fmt_gene_value(row.get("Q0")),
        _fmt_gene_value(row.get("Eacc_MVpm")),
        _fmt_gene_value(row.get("L_m")),
        _fmt_gene_value(row.get("beta")),
        _fmt_gene_value(row.get("pulse_length_ns")),
        _fmt_gene_value(row.get("rep_rate_Hz")),
        _fmt_gene_value(row.get("P_aux_kW")),
        _fmt_gene_value(row.get("source_power_avail_kW")),
        _fmt_gene_value(row.get("cooling_capacity_kW")),
        _fmt_gene_value(row.get("surf_factor")),
        _fmt_gene_value(row.get("geom_factor")),
        _fmt_gene_value(row.get("freq_factor")),
        _fmt_gene_value(row.get("fab_sigma_um")),
        _fmt_gene_value(row.get("delta_allow_um")),
        _fmt_gene_value(row.get("S_f")),
        _fmt_gene_value(row.get("S_phi")),
        _fmt_gene_value(row.get("S_c")),
    ]
    blob = "|".join(fields).encode("utf-8")
    return "cand_" + hashlib.sha256(blob).hexdigest()[:16]