daniel-koehn / DENISE-Black-Edition

2D time-domain isotropic (visco)elastic FD modeling and full waveform inversion (FWI) code for P/SV-waves
GNU General Public License v2.0
121 stars 66 forks source link

SegFault running example notebook and Python in general #49

Open berniez opened 8 months ago

berniez commented 8 months ago

Hello I'm new to the world of geophysics, and was trying to use DENISE as my first FWI tool. When trying to run the example notebook, the forward modeling works just fine, but when doing FWI proper I get a segmentation fault error when calculating the gradients.

PE 0 is writing model to taper/taper_coeff_hor.bin.0.0

The error (part of it) is in the image attached. I get the same thing when running a simple model I created, no matter if from the command line or inside a notebook. Setting SWS_TAPER_GRAD_HOR=0 gives the same error in a later step, after

writing merged model file to /home/delta/bernardo/TestFWI_Denise/marmousi_tutorial/jacobian/gradient_Test_p_u.old Opening model files: /home/delta/bernardo/TestFWI_Denise/marmousi_tutorial/jacobian/gradient_Test_p_u.old.??? ... finished. Copying... ... finished. Use ximage n1=174 < /home/delta/bernardo/TestFWI_Denise/marmousi_tutorial/jacobian/gradient_Test_p_u.old label1=Y label2=X title=/home/delta/bernardo/TestFWI_Denise/marmousi_tutorial/jacobian/gradient_Test_p_u.old to visualize model.

Interestingly, running the steps listed in quickstart_MARMOUSI.txt works with no issues.

Any help is appreciated

image