seistorch/examples/inversion/misfits/l1/forward.yml at dev · StarrMoonn/seistorch · GitHub

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
# Random seed for pytorch and all shuffle functions
seed: 20230503

name: forward_modeling

# Data type to use for tensors, either float32 or float64
dtype: float32

equation: acoustic_habc

training:

  implicit:
    use: false
    pretrained:

  minibatch: true

  batch_size: 20

  # Number of training epochs
  N_epochs: 50

  # Optimizer learning rate
  lr: 0.001

  # Decay the learning rate by each of multi-scale
  scale_decay: 0.85

  # decay the lr by each of epochs
  lr_decay: 0.995

  # Filter order
  filter_ord: 3

geom:

  # The observer data
  obsPath: ./observed.npy

  # Wave speed model from disk_file for forward modeling
  truePath:
    vp: ../../../models/marmousi_model/true_vp.npy
    vs:
    rho:
    Q:


  sources: ./geometry/sources.pkl

  receivers: ./geometry/receivers.pkl

  seabed: ../../../models/marmousi_model/seabed.npy

  wavelet:

  multiple: false

  boundary_saving: true

  wavelet_delay: 256

  wavelet_inverse: false
  #./geometry/ricker_lowcut5hz.npy

  source_type:
    - h1

  receiver_type:
    - h1

  # Inital Wave speed model from disk_file
  initPath:
    vp: ../../../models/marmousi_model/linear_vp.npy
    vs:
    rho:
    Q:

  invlist:
    vp: true
    vs:
    rho:
    q:

  # Save the results of FWI
  inv_savePath:

  multiscale:
    - 1.0
    - 3.0
    - 5.0
    - 8.0
    - all

  # Time Sampling step
  dt: 0.002

  # Time samples
  nt: 2000

  # Main frequency of ricker wave
  fm: 8.0

  # Spatial grid step size
  h: 20.0

  # Total Shots
  Nshots: 999

  boundary:
    type: habc
    width: 50    # PML thickness in number of grid cells