-
Notifications
You must be signed in to change notification settings - Fork 0
/
0-create_data_LHS.py
131 lines (106 loc) · 3.54 KB
/
0-create_data_LHS.py
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
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Author: charles
# @Date: 2021-01-24 20:01:11
# @Last modified by: charles
# @Last modified time: 2022-08-05 15:08:83
import torch
import numpy as np
from tqdm import tqdm
from SALib.sample import latin
import matplotlib.pyplot as plt
from ppip_model import forward_spherical, mixture
from utilities import normalize, polarize, log_complex
import plotlib
norms = ['raw', 'log', 'norm', 'pv']
f = np.logspace(2, 6, 32)
w = 2*np.pi*f
problem = {
'num_vars': 8,
'names': list(mixture.keys()),
'bounds': [[np.log10(1e-5), np.log10(1e-3)],
[np.log10(1e-3), np.log10(2e-1)],
[-7, -5],
[0, 5],
# [5, 25],
[-11, -9],
[-10, -8],
[-3, 0],
# [70, 90],
[-11, -9],
],
}
n_ex = 100000
param_values = latin.sample(problem, n_ex)
Z = np.empty((param_values.shape[0], w.shape[0], 2))
for i, X in enumerate(tqdm(param_values)):
f = forward_spherical(w, *X)
Z[i, :, 0] = f.real
Z[i, :, 1] = f.imag
Z_0 = forward_spherical(w, **mixture)
Z_0 = torch.view_as_real(torch.from_numpy(Z_0))
Z_0 = torch.unsqueeze(Z_0, 0).numpy()
Z = np.append(Z, Z_0, axis=0)
# Z += Z*1e-2*np.random.randn(*Z.shape)
Zn = {}
Zn_0 = {}
for norm in norms:
if norm == 'raw':
Zn[norm] = Z
elif norm == 'log':
Zn[norm] = np.log10(Z)
elif norm == 'sqrt':
Zn[norm] = np.sqrt(Z)
elif norm == 'cbrt':
Zn[norm] = np.cbrt(Z)
elif norm == 'pv':
Zn[norm] = log_complex(Z)
elif norm == 'rec':
Zn[norm] = 1/Z
elif norm == 'polar':
Zn[norm] = polarize(Z)
elif norm == 'norm':
Zn[norm] = normalize(Z, axis=1)
# Zn[norm] += 0.001*Zn[norm]*np.random.randn(*Zn[norm].shape)
Zn_0[norm] = np.expand_dims(Zn[norm][-1], axis=0).copy()
data = torch.tensor(Zn_0[norm]).float()
data_fpath = f'./data/mixture-{norm}.pt'
torch.save(data, data_fpath)
Zn[norm] = np.delete(Zn[norm], -1, axis=0)
data = torch.tensor(Zn[norm]).float()
data_fpath = f'./data/dataset-{norm}.pt'
torch.save(data, data_fpath)
param_values = torch.tensor(param_values).float()
param_fpath = './data/parameters.pt'
torch.save(param_values, param_fpath)
f = np.logspace(2, 6, 32)
legends = [
[None, None, r"$\sigma'$ (S/m)", r"$\sigma''$ (S/m)"],
# [None for _ in range(4)],
# [None for _ in range(4)],
# [None for _ in range(4)],
[None, None, r"$\log_{10}\sigma'$", r"$\log_{10}\sigma''$"],
# [None, None, r"$\sqrt{\sigma'}$", r"$\sqrt{\sigma''}$"],
[None, None, r"$\sigma'_\mathrm{norm}$", r"$\sigma''_\mathrm{norm}$"],
[None, None, r"$\ln\vert\sigma_\mathrm{eff}\vert$", r"$\varphi$ (rad)"]
]
colors = ['C0', 'C3', 'C1', 'C2']
fig, axs = plt.subplots(2, 2, figsize=(1.3*3.5433, 3.5433), sharex=True)
for i, norm in enumerate(norms):
kwargs = dict(s=5, color=colors[i])
ax = axs.flat[i]
ax.scatter(f, Zn_0[norm][0][:, 0], marker='^', label=legends[i][2],
**kwargs)
ax.scatter(f, Zn_0[norm][0][:, 1], marker='v', label=legends[i][3],
**kwargs)
ax.set_title(norm)
ax.legend(fontsize='small')
ax.set_xscale('log')
ax.set_xlim([f.min(), f.max()])
# Set common labels
fig.text(0.5, 0.0, 'Frequency (Hz)', ha='center', va='center')
fig.text(0.0, 0.5, 'Effective complex conductivity', ha='center', va='center',
rotation='vertical')
plt.tight_layout()
for ext in ['.png', '.pdf']:
plt.savefig(f'./figures/default-mixture-transforms{ext}')