Difference between revisions of "Python interface"
Line 140: | Line 140: | ||
s.fileSave_2("/tmp/dc.page") | s.fileSave_2("/tmp/dc.page") | ||
s.exportImage_2("/tmp/dc.pdf") | s.exportImage_2("/tmp/dc.pdf") | ||
+ | |||
+ | == Other examples == | ||
+ | |||
+ | * [[Python example: editing HV results]] |
Latest revision as of 08:42, 22 October 2024
Some functionalities are accessible from a Python script. Currently (from version 3.5.0) there are two modules:
- GeopsyPyCoreWave: surface wave computation
- GeopsyPyScifigs: graphical tools to plot data (compatible with Qt 6.5 and above)
Other modules will be probably proposed in the future (e.g. Neighborhood inversion). The very basic interface may quickly evolve according to user feedback (forum)
Contents
Installation
The modules are compiled with Geopsy package if Python include files are in the include paths. At configure step, add Python include path (among other options not listed here, see general instructions specific to your platform):
./configure -I /usr/include/python3.9
numpy must be installed and its include files must be available during compilation. If the configuration fails complaining about Python, search your base tree to get the path to "Python.h" and eventually "numpy/arrayobject.h":
find /usr /opt -name "Python.h" find /usr /opt -name "arrayobject.h"
On some platforms, the directory containing "Python.h" has symbolic links to the include directories of the modules. It is thus not necessary to specify the include paths of the modules in the configuration options.
Currently, modules are tested under Linux only. Under Windows and Mac OS, GeopsyPySciFigs is probably currently not working.
Upon completion of the build process, the modules are dynamic libraries located in the lib directory of the install directory. Their names start with "lib" which prevents the recognition by Python. For releases before 3.5.3, we suggest to create a new directory at the same level as lib and to modify the library names as follow:
mkdir python cp lib/libGeopsyPyCoreWave.so python/GeopsyPyCoreWave.so cp lib/libGeopsyPySciFigs.so python/GeopsyPySciFigs.so
This step is included in "make install" from release 3.5.3. Finally, you have to add the path to Python
export PYTHONPATH=/path/where/geopsy/is/installed/python:$PYTHONPATH
Interface documentation
The list of available functions in each module is provided with the help function. Their required arguments are also shortly described.
import GeopsyPyCoreWave as gp help(gp)
Example
In this section, we describe a small example (SciFigs-example-01.py) to compute dispersion and ellipticity curves, and to plot them on the screen, in a .page file and in a PDF file. The figure on the right is a SVG export with option print contents as bitmap unchecked.
Running the example
To run the Python script under Linux, first make it executable:
$ chmod a+x SciFigs-example-01.py $ export PYTHONPATH=... see previous section ...
Before running the script, check the first line and eventually modify the path to Python binary and the version of Python. By default it is
#!/usr/bin/python3.9 -i
but it could be also
#!/opt/conda/bin/python3 -i
When this is done, you can run it
$ ./SciFigs-example-01.py
Description
First import of the required modules:
import GeopsyPySciFigs as sf import numpy as np import GeopsyPyCoreWave as gp
Construction of the NumPY arrays which define the 1D layered model and the frequency vector where to compute the curves.
nmodes=7 freq=np.logspace(np.log10(1), np.log10(50), 500) omega=freq.copy() omega*=2*np.pi h=np.array([10,20]) vp=np.array([500,1000,3000]) vs=np.array([200,600,1500]) rho=np.array([2000,2000,2500])
Computation of dispersion and ellipticity curves.
slowRayleigh=gp.rayleighDispersionCurve(nmodes, 0, h, vp, vs, rho, omega) slowRayleighGroup=gp.rayleighDispersionCurve(nmodes, 1, h, vp, vs, rho, omega) ellRayleigh=gp.rayleighEllipticityCurve(nmodes, h, vp, vs, rho, omega) ellRayleigh=np.arctan(ellRayleigh)*(180/np.pi) slowLove=gp.loveDispersionCurve(nmodes, 0, h, vs, rho, omega)
Definition of useful functions to set plot attributes.
def layoutAttributes(p, x, y, w, h): p.setAnchor("TopRight") p.xAxis().setSizeInfo(w) p.yAxis().setSizeInfo(h) p.setPrintX(x) p.setPrintY(y)
def frequencyAttributes(p): p.xAxis().setMin(1) p.xAxis().setMax(50) p.xAxis().setScaleType("Log") p.xAxis().setTitle("Frequency (Hz)")
def dispersionAttributes(p): frequencyAttributes(p) p.yAxis().setScaleType("InverseLog") p.yAxis().setMin(1/1500) p.yAxis().setMax(1/150)
Creation of a blank GraphicSheet.
s=sf.newSheet()
Add a dispersion curve plot with Rayleigh phase velocity.
prp=sf.newPlot(s) sf.addCurves(prp, freq, slowRayleigh) layoutAttributes(prp, 12, 1, 11 ,6) dispersionAttributes(prp) prp.yAxis().setTitle("Rayleigh phase velocity (m/s)")
Add a dispersion curve plot with Rayleigh group velocity.
prg=sf.newPlot(s) sf.addCurves(prg, freq, slowRayleighGroup) layoutAttributes(prg, 23, 1, 11 ,6) dispersionAttributes(prg) prg.yAxis().setTitle("Rayleigh group velocity (m/s)")
Add an ellipticity curve plot.
per=sf.newPlot(s) sf.addCurves(per, freq, ellRayleigh) layoutAttributes(per, 12, 7, 11 ,6) frequencyAttributes(per) per.yAxis().setTitle("Angular ellipticity (deg)") per.yAxis().setMin(-90) per.yAxis().setMax(90) per.graphContents().layer(0).setSignThreshold(45)
Add a dispersion curve with Love phase velocity
pl=sf.newPlot(s) sf.addCurves(pl, freq, slowLove) layoutAttributes(pl, 23, 7, 11 ,6) dispersionAttributes(pl) pl.yAxis().setTitle("Love phase velocity (m/s)")
Output results in .page file and in a PDF.
s.setPaperOrientation("Landscape") s.fileSave_2("/tmp/dc.page") s.exportImage_2("/tmp/dc.pdf")