SimPEG has been slowly included into mtpy. The most basic inversion is 1D. Here, a few different methods for inverting data in 1D are presented.
from pathlib import Path
import numpy as np
from matplotlib import pyplot as plt
from mtpy import MTCollectionfrom mtpy import MTCollection
mc = MTCollection()
mc.open_collection(Path(r"c:\Users\jpeacock\OneDrive - DOI\SAGE\2024\EDI_Files\SAGE_tf_all.h5"))
#mc.open_collection(Path(r"../../../scratch/nm05/jp0633/musgraves_edis/musgraves_tf.h5"))
from ipywidgets import widgets, interact
station_names = mc.dataframe.station.values
def foo(name, component):
tf = mc.get_tf(name)
tf.plot_mt_response()
Q = interact(
foo,
name=widgets.Select(options=station_names, value='V2404'),
component=widgets.RadioButtons(options=['xy', 'yx', 'det'], value='det')
)Loading...
name = Q.widget.kwargs['name']
component = Q.widget.kwargs['component']
tf = mc.get_tf(name)
fig, ax = plt.subplots(1,1, figsize=(5,5))
mc.dataframe.plot(x='longitude', y='latitude', marker='.', linestyle='None', ax=ax)
ax.plot(tf.longitude, tf.latitude, 'ro')
ax.set_title(name)24:06:28T10:32:15 | WARNING | line:300 |mtpy.core.mt_collection | get_tf | Found multiple transfer functions with ID V2404. Suggest setting survey, otherwise returning the TF from survey SAGE.
inv_obj = tf.to_simpeg_1d(mode="det")24:06:28T10:32:55 | INFO | line:1109 |mtpy.core.mt | to_simpeg_1d | Using default errors for impedance
C:\Users\jpeacock\OneDrive - DOI\Documents\GitHub\mtpy-v2\mtpy\modeling\simpeg\recipes\inversion_1d.py:194: RuntimeWarning: divide by zero encountered in log10
np.where(np.log10(abs(np.diff(sub_df.res))) > max_diff_res)[0] + 1
SimPEG.InvProblem is setting bfgsH0 to the inverse of the eval2Deriv.
***Done using same Solver, and solver_opts as the Simulation1DRecursive problem***
model has any nan: 0
============================ Inexact Gauss Newton ============================
# beta phi_d phi_m f |proj(x-g)-x| LS Comment
-----------------------------------------------------------------------------
x0 has any nan: 0
0 7.23e+02 1.63e+03 0.00e+00 1.63e+03 1.02e+03 0
1 3.62e+02 2.05e+02 1.47e-04 2.05e+02 1.65e+02 0
C:\Users\jpeacock\Anaconda3\envs\py39\lib\site-packages\SimPEG\electromagnetics\natural_source\simulation_1d.py:129: RuntimeWarning: overflow encountered in tanh
tanhs = np.tanh(alphas[:-1] * thicknesses[:, None])
C:\Users\jpeacock\Anaconda3\envs\py39\lib\site-packages\SimPEG\electromagnetics\natural_source\simulation_1d.py:171: RuntimeWarning: overflow encountered in tanh
tanhs = np.tanh(alphas[:-1] * thicknesses[:, None])
------------------------- STOP! -------------------------
1 : |fc-fOld| = 0.0000e+00 <= tolF*(1+|f0|) = 1.6342e+02
0 : |xc-x_last| = 6.2437e+00 <= tolX*(1+|x0|) = 3.3887e+00
0 : |proj(x-g)-x| = 1.6511e+02 <= tolG = 1.0000e-01
0 : |proj(x-g)-x| = 1.6511e+02 <= 1e3*eps = 1.0000e-02
0 : maxIter = 40 <= iter = 2
------------------------- DONE! -------------------------
tf.compute_model_z_errors(error_value=3, error_type="data")
inv_obj = tf.to_simpeg_1d(mode="det", **{"p_s": 2, "p_z": 0, "use_irls": True})C:\Users\jpeacock\OneDrive - DOI\Documents\GitHub\mtpy-v2\mtpy\modeling\simpeg\recipes\inversion_1d.py:194: RuntimeWarning: divide by zero encountered in log10
np.where(np.log10(abs(np.diff(sub_df.res))) > max_diff_res)[0] + 1
C:\Users\jpeacock\Anaconda3\envs\py39\lib\site-packages\SimPEG\directives\directives.py:337: UserWarning: Without a Linear preconditioner, convergence may be slow. Consider adding `Directives.UpdatePreconditioner` to your directives list
[directive.validate(self) for directive in self.dList]
SimPEG.InvProblem is setting bfgsH0 to the inverse of the eval2Deriv.
***Done using same Solver, and solver_opts as the Simulation1DRecursive problem***
model has any nan: 0
============================ Inexact Gauss Newton ============================
# beta phi_d phi_m f |proj(x-g)-x| LS Comment
-----------------------------------------------------------------------------
x0 has any nan: 0
0 3.66e+09 4.63e+09 0.00e+00 4.63e+09 4.49e+09 0
1 1.83e+09 3.46e+08 1.45e-06 3.46e+08 6.26e+08 0
2 9.14e+08 1.79e+07 1.11e-03 1.89e+07 9.45e+07 0 Skip BFGS
3 4.57e+08 5.92e+06 1.65e-03 6.68e+06 1.18e+07 0 Skip BFGS
4 2.28e+08 1.55e+06 3.13e-03 2.26e+06 1.44e+07 0 Skip BFGS
5 1.14e+08 1.06e+06 3.10e-03 1.41e+06 9.04e+05 0
6 5.71e+07 6.18e+05 4.14e-03 8.54e+05 1.58e+06 0
7 2.86e+07 5.84e+05 4.46e-03 7.11e+05 1.45e+05 0 Skip BFGS
8 1.43e+07 5.55e+05 5.19e-03 6.29e+05 2.12e+05 0 Skip BFGS
9 7.14e+06 5.26e+05 6.69e-03 5.73e+05 5.91e+05 0 Skip BFGS
10 3.57e+06 4.84e+05 1.12e-02 5.24e+05 1.09e+06 0 Skip BFGS
11 1.78e+06 4.81e+05 1.12e-02 5.01e+05 1.13e+05 0
12 8.92e+05 4.47e+05 1.77e-02 4.63e+05 9.93e+05 2
13 4.46e+05 4.25e+05 3.05e-02 4.39e+05 2.97e+06 2 Skip BFGS
14 2.23e+05 3.98e+05 4.83e-02 4.08e+05 3.57e+06 2 Skip BFGS
15 1.12e+05 3.64e+05 5.29e-02 3.69e+05 3.49e+05 0
16 5.58e+04 3.46e+05 1.02e-01 3.52e+05 1.03e+06 2
C:\Users\jpeacock\Anaconda3\envs\py39\lib\site-packages\SimPEG\electromagnetics\natural_source\simulation_1d.py:129: RuntimeWarning: overflow encountered in tanh
tanhs = np.tanh(alphas[:-1] * thicknesses[:, None])
17 2.79e+04 3.36e+05 1.24e-01 3.39e+05 1.79e+06 3
18 1.39e+04 3.31e+05 1.41e-01 3.33e+05 2.51e+06 4 Skip BFGS
C:\Users\jpeacock\Anaconda3\envs\py39\lib\site-packages\SimPEG\electromagnetics\natural_source\simulation_1d.py:171: RuntimeWarning: overflow encountered in tanh
tanhs = np.tanh(alphas[:-1] * thicknesses[:, None])
19 6.97e+03 3.24e+05 1.78e-01 3.25e+05 3.65e+06 1
20 3.49e+03 3.20e+05 1.89e-01 3.21e+05 3.81e+06 4
21 1.74e+03 3.09e+05 2.98e-01 3.10e+05 1.67e+06 1 Skip BFGS
22 8.71e+02 3.07e+05 3.44e-01 3.07e+05 2.13e+06 4 Skip BFGS
23 4.36e+02 2.88e+05 2.77e-01 2.88e+05 1.27e+06 0
24 2.18e+02 2.87e+05 3.80e-01 2.87e+05 2.02e+06 3
25 1.09e+02 2.87e+05 3.56e-01 2.87e+05 2.16e+06 3
26 5.45e+01 2.85e+05 3.90e-01 2.85e+05 1.66e+06 0
27 2.72e+01 2.85e+05 4.00e-01 2.85e+05 1.97e+06 4
28 1.36e+01 2.81e+05 5.01e-01 2.81e+05 1.31e+06 0 Skip BFGS
29 6.81e+00 2.80e+05 5.79e-01 2.80e+05 1.80e+06 2 Skip BFGS
30 3.40e+00 2.76e+05 5.63e-01 2.76e+05 2.59e+05 0
31 1.70e+00 2.75e+05 6.22e-01 2.76e+05 6.24e+05 2 Skip BFGS
32 8.51e-01 2.75e+05 6.42e-01 2.75e+05 1.97e+05 0
33 4.26e-01 2.74e+05 6.15e-01 2.74e+05 6.39e+05 3 Skip BFGS
34 2.13e-01 2.74e+05 7.08e-01 2.74e+05 1.12e+06 4
35 1.06e-01 2.74e+05 7.12e-01 2.74e+05 1.20e+06 5
36 5.32e-02 2.74e+05 7.14e-01 2.74e+05 1.13e+06 0
37 2.66e-02 2.74e+05 6.76e-01 2.74e+05 1.16e+06 5 Skip BFGS
38 1.33e-02 2.74e+05 5.92e-01 2.74e+05 1.13e+06 3
39 6.65e-03 2.72e+05 6.92e-01 2.72e+05 3.66e+05 0
40 3.32e-03 2.71e+05 9.84e-01 2.71e+05 6.47e+05 4 Skip BFGS
------------------------- STOP! -------------------------
1 : |fc-fOld| = 8.9766e+02 <= tolF*(1+|f0|) = 4.6292e+08
0 : |xc-x_last| = 2.3275e+01 <= tolX*(1+|x0|) = 3.3887e+00
0 : |proj(x-g)-x| = 6.4738e+05 <= tolG = 1.0000e-01
0 : |proj(x-g)-x| = 6.4738e+05 <= 1e3*eps = 1.0000e-02
1 : maxIter = 40 <= iter = 40
------------------------- DONE! -------------------------
inv_obj.plot_response(7)
inv_obj = tf.to_simpeg_1d(mode="det", **{"p_s": 0, "p_z": 0, "use_irls": True})C:\Users\jpeacock\OneDrive - DOI\Documents\GitHub\mtpy-v2\mtpy\modeling\simpeg\recipes\inversion_1d.py:194: RuntimeWarning: divide by zero encountered in log10
np.where(np.log10(abs(np.diff(sub_df.res))) > max_diff_res)[0] + 1
C:\Users\jpeacock\Anaconda3\envs\py39\lib\site-packages\SimPEG\directives\directives.py:337: UserWarning: Without a Linear preconditioner, convergence may be slow. Consider adding `Directives.UpdatePreconditioner` to your directives list
[directive.validate(self) for directive in self.dList]
SimPEG.InvProblem is setting bfgsH0 to the inverse of the eval2Deriv.
***Done using same Solver, and solver_opts as the Simulation1DRecursive problem***
model has any nan: 0
============================ Inexact Gauss Newton ============================
# beta phi_d phi_m f |proj(x-g)-x| LS Comment
-----------------------------------------------------------------------------
x0 has any nan: 0
0 3.80e+09 4.63e+09 0.00e+00 4.63e+09 4.49e+09 0
1 1.90e+09 3.46e+08 1.45e-06 3.46e+08 6.26e+08 0
2 9.49e+08 1.79e+07 1.11e-03 1.89e+07 9.45e+07 0 Skip BFGS
3 4.74e+08 5.92e+06 1.65e-03 6.70e+06 1.18e+07 0 Skip BFGS
4 2.37e+08 1.55e+06 3.09e-03 2.28e+06 1.43e+07 0 Skip BFGS
5 1.19e+08 1.07e+06 3.07e-03 1.44e+06 9.16e+05 0
6 5.93e+07 6.21e+05 4.12e-03 8.65e+05 1.62e+06 0
7 2.97e+07 5.89e+05 4.38e-03 7.19e+05 2.65e+05 0 Skip BFGS
8 1.48e+07 5.57e+05 5.12e-03 6.33e+05 2.19e+05 0 Skip BFGS
9 7.41e+06 5.22e+05 6.88e-03 5.73e+05 6.07e+05 0 Skip BFGS
10 3.71e+06 4.83e+05 1.09e-02 5.23e+05 1.09e+06 0 Skip BFGS
11 1.85e+06 4.80e+05 1.08e-02 5.01e+05 1.15e+05 0
12 9.27e+05 4.48e+05 1.68e-02 4.64e+05 7.98e+05 2
C:\Users\jpeacock\Anaconda3\envs\py39\lib\site-packages\SimPEG\electromagnetics\natural_source\simulation_1d.py:129: RuntimeWarning: overflow encountered in tanh
tanhs = np.tanh(alphas[:-1] * thicknesses[:, None])
13 4.63e+05 4.24e+05 2.91e-02 4.38e+05 2.83e+06 2 Skip BFGS
14 2.32e+05 4.10e+05 3.62e-02 4.19e+05 3.28e+06 3 Skip BFGS
15 1.16e+05 4.00e+05 6.45e-02 4.08e+05 1.52e+06 2 Skip BFGS
16 5.79e+04 3.93e+05 1.12e-01 4.00e+05 1.80e+06 2
17 2.90e+04 3.40e+05 2.24e-01 3.47e+05 5.21e+06 0 Skip BFGS
18 1.45e+04 2.96e+05 2.23e-01 2.99e+05 3.97e+05 0
19 7.24e+03 2.91e+05 2.69e-01 2.93e+05 1.69e+06 3
20 3.62e+03 2.86e+05 2.80e-01 2.87e+05 1.22e+05 0
21 1.81e+03 2.82e+05 3.38e-01 2.83e+05 1.62e+06 3 Skip BFGS
C:\Users\jpeacock\Anaconda3\envs\py39\lib\site-packages\SimPEG\electromagnetics\natural_source\simulation_1d.py:171: RuntimeWarning: overflow encountered in tanh
tanhs = np.tanh(alphas[:-1] * thicknesses[:, None])
22 9.05e+02 2.78e+05 4.29e-01 2.79e+05 1.66e+06 2 Skip BFGS
23 4.53e+02 2.78e+05 5.13e-01 2.78e+05 1.74e+06 3 Skip BFGS
24 2.26e+02 2.70e+05 5.01e-01 2.70e+05 1.06e+05 0
25 1.13e+02 2.69e+05 5.83e-01 2.69e+05 1.55e+06 3
26 5.66e+01 2.67e+05 5.97e-01 2.67e+05 1.56e+05 0
27 2.83e+01 2.67e+05 6.16e-01 2.67e+05 4.16e+05 6 Skip BFGS
28 1.41e+01 2.65e+05 7.25e-01 2.65e+05 1.10e+06 3 Skip BFGS
29 7.07e+00 2.65e+05 8.31e-01 2.65e+05 1.48e+06 3 Skip BFGS
30 3.54e+00 2.63e+05 7.98e-01 2.63e+05 2.25e+05 0
31 1.77e+00 2.63e+05 8.48e-01 2.63e+05 5.58e+05 5
32 8.84e-01 2.62e+05 8.85e-01 2.62e+05 7.24e+05 5 Skip BFGS
33 4.42e-01 2.62e+05 8.89e-01 2.62e+05 7.45e+05 9
34 2.21e-01 2.62e+05 9.16e-01 2.62e+05 7.44e+05 5
35 1.10e-01 2.62e+05 8.93e-01 2.62e+05 7.47e+05 4
36 5.52e-02 2.62e+05 9.19e-01 2.62e+05 7.58e+05 4 Skip BFGS
37 2.76e-02 2.62e+05 9.21e-01 2.62e+05 7.63e+05 5
38 1.38e-02 2.62e+05 9.29e-01 2.62e+05 7.66e+05 9
39 6.90e-03 2.62e+05 9.84e-01 2.62e+05 7.78e+05 7 Skip BFGS
40 3.45e-03 2.62e+05 9.62e-01 2.62e+05 7.94e+05 4
------------------------- STOP! -------------------------
1 : |fc-fOld| = 3.4856e+00 <= tolF*(1+|f0|) = 4.6292e+08
1 : |xc-x_last| = 1.5990e+00 <= tolX*(1+|x0|) = 3.3887e+00
0 : |proj(x-g)-x| = 7.9374e+05 <= tolG = 1.0000e-01
0 : |proj(x-g)-x| = 7.9374e+05 <= 1e3*eps = 1.0000e-02
1 : maxIter = 40 <= iter = 40
------------------------- DONE! -------------------------
inv_obj.plot_response(8)
mc.close_collection()24:06:28T10:38:07 | INFO | line:759 |mth5.mth5 | close_mth5 | Flushing and closing c:\Users\jpeacock\OneDrive - DOI\SAGE\2024\EDI_Files\SAGE_tf_all.h5