Activity for Max Moorkamp
-
Max Moorkamp committed [r1342] on jif3D
Made home path more portable
-
Fixed final output when cross-gradient coupling ist used
-
updated outdated README file
-
Removed outdated --mutual_information option from options string, updated colormap for density model to make it visually distinct
-
-
Removed depreceated and unused include statement from FileUtil
-
-
-
Joint inversion now shows information how it reads coupling validity
-
Improved output of histograms for VI
-
Improved messages when setting up Gravity and Magnetic data
-
Made some changes for compatibility with latest version of HPX, update to singularity/apptainer defintion file
-
Added namespace specifications for some functions that were accepted for older versions of g++ but fail in version 14
-
Added a definition file to build Singularity/Apptainer Containers
-
Increased maximum number of ray segments before ray is considered lost.
-
joint inversion now outputs hit count per cell for tomography
-
Improved handling of invalid settings and values, can now specify minimum and maximum values for binning in VI
-
Rotation angles in tipper file are optional when reading in data
-
Can now specify a reference model for regularization for the each method in the joint inversion, can also just use the starting model as a reference model
-
fixed typo in options that prevented slowness covariance from being read in properly
-
Added missing override to SetupOptions definition in some setup classes
-
Changed threshold for merging layers in x3d to avoid problems with homogeneous starting models
-
Updated tutorials to reflect new use of pyproj library
-
-
-
renamed mutual information to VI to better reflect what is actually calculated and reduce confusion between description in papers and code
-
renamed mutual information to VI to better reflect what is actually calculated and reduce confusion between description in papers and code
-
Optimized computation of VI constraint, needs some testing
-
Added check to joint inversion if measurement sirtes are within modelling domain to warn user of possible problems
-
Modified covarianxce setup to account for non-grid parameters and stochastic covariance interactions
-
Added profiling program forVI contraint
-
Added a very simple 1D MT inversion code for demonstration purposes
-
Some clean up of code to reduce compiler warnings and number of failed tests
-
Added nectdf-cxx4 to directory structure for easier compilation and less external code dependencies. Some cleanup of include statements associated with netcdf
-
Joint inversion now outputs synthetic tipper data
-
Changed transform for tipper so that distortion parameters are not passed along any more
-
The 1D responses now get written out for comparison
-
Added compilation instructions for file compile
-
added source code of x3d so users can compile themselves
-
surface wave inversions outputs station locations
-
fixed setup of dc resistivity inversion to consider model covariance
-
changed filename for output of dispersion tests to avoid overwriting other files
-
added more meaningful error message when trying to read variable from netcdf file
-
fixed setting covariance when magnetic model is used as a constrained
-
added option to use identical density values for gravity and surface wave modelling in joint inversion
-
- added possibility to invert for absolute densities
-
- added possibility to invert for absolute densities
-
changed setup of MT when conductivity model is used as a constraint to ensure model values are copied
-
-
fixed cell position calculation in tomography calculator, refined tests for surface wave dispersion forward and gradient, tests look self consistent now
-
clean up and comments for SurfaceWave calculation
-
the ele field in the tomography ray calculatation is now set to usable values
-
Changed default velocity range for surface waves to be consistent with model units
-
some optimization to Surface wave calculation, should use less memory now
-
fixed corner case in vtk test
-
improved handling of density and pvel velocity in surface wave inversion
-
added test program to minimize variation of information between two models only
-
more work on surface wave inversion with ray tracing, made extra transform to calculate density and P-velocity from s-velocity
-
more work on data covariance setup, needs more checking and some cleaning
-
-
-
Added pragmas to suppress warnings from library code
-
-
-
-
-
continued implementation of ray tracing, needs further refinement and testing
-
started implementation of ray tracing in surface wave calculator, currently still incomplete
-
Added model origin information to first arrival travel time calculator so raypaths can be plotted correctly
-
refactored hpx gravity calculation to make sure computations overlap properly and sensitivities are stored properly
-
Added a Tutorial that demonstrates the use of topography and other prior information when jointly inverting gravity and magnetic data
-
Tear specification for regularization in new jointinv is now transferred to cross-gradient again
-
Fixed handling of stochastic covariance for non-grid parameters
-
added to fstream to include in field calculator to avoid errors in gcc 12
-
Fixed setup of covariance vector when not all data types are present, also added some progress info for users
-
Changed initialization of MT part of joint inversion to ensure Distortion correction works
-
Fixed normalization factor for MI inversion when specifying a validity domain
-
Fixed units and gradient calculation for magnetization inversion, added test for forward (comparison with SIMPEG) and gradient calculation
-
made some preparations for providing Docker version of jif3D
-
-
improved error checking and output of misfit for MT data
-
fixed some hpx related warnings in tomography code
-
- Magnetics is now CUDA enabled and tested to give correct results
-
added error checking to CUDA computations to guad against erroneous results when compiling with wrong version
-
-
-
Added first CUDA calculation for total field magnetics and updated associated tests for magnetics and gravity
-
Gravity and Magnetic setup now give an indication of estimated disk space, further fixes to constraint inversion with conductivity model
-
fixed setup when conductivity model is used as a constraint
-
some changes to includes for gravity calculation to adjust to latest version of hpx (1.8.0)
-
correction to order of parameters in Tutorials
-
Updated Tutorials to consider DCResistivity option in jointinv, made handling of covariance consistent across methods and constrained inversions more convenient
-
Continued overhaul of joint inversion main program
-
Added distortion correction to new flexible version of joint inversion, still needs some testing
-
updated Tutorials to new grav_jointinv program, this might need further adjustments in the near future
-
Significant overhaul of the way the joint inversion is set up. This is a big step towards allowing arbitrary combinations of data to be jointly inverted and getting rid of different programs for different sub-types of data. However, this results in breaking changes to the user interface, i.e. the order in which information is input has changed compared to previous versions. Also, there is likely going to be some flux in the interface while this is being refined.
-
Updated installation instructions
-
some adjustment to code formatting and added xml reading to convertmtt
-
changed formatting for output of JointObjective to deal with longer names for terms of the objective function
-
Implemented ExtraParameterSetter and copy operator for MagnetizationModel so that the inversion can deal with the fact that there are three values for each model cell
1 >