Opening .report file

This forum is dedicated to discuss all problems and suggestions related to the inversion software
Post Reply
Aniket19061990
Posts: 13
Joined: Sun Jan 24, 2021 11:11 am

Opening .report file

Post by Aniket19061990 »

Hi.
With which software can .report file be opened?
I ran inversion and run_01.report file was generated of size about 79000 kb. However, by mistake, I again ran run1 and the file got overwritten, now having size of around 159500 kb. How can I check whether the results of earlier inversion are also present there or it is completely overwritten with only the second inversion results?
admin
Site Admin
Posts: 822
Joined: Mon Aug 13, 2007 11:48 am
Location: ISTerre
Contact:

Re: Opening .report file

Post by admin »

Hi,

You can inspect these files with gpdcreport and gpviewdcreport. If it is overwritten, I'm afraid that the first run is lost. In principle with the parameters and the target, you can re-generate it inside dinver.

Best regards,

Marc
Aniket19061990
Posts: 13
Joined: Sun Jan 24, 2021 11:11 am

Re: Opening .report file

Post by Aniket19061990 »

Thank you for the prompt reply. I could find it. Actually the Vs profiles generated from both are there (60000(Ns0+Ns) was there in each case and in this I got 120174 Vs profiles). However, I want to separate them out because I want lowest misfit profile from each of them separately. Is it possible to make two separate .report files from this?
Last edited by Aniket19061990 on Wed Nov 17, 2021 5:36 pm, edited 1 time in total.
admin
Site Admin
Posts: 822
Joined: Mon Aug 13, 2007 11:48 am
Location: ISTerre
Contact:

Re: Opening .report file

Post by admin »

To get only first part it is relatively easy. Let's imagine that there are 10000 models in the original.report:

Code: Select all

$ gpdcreport original.report -count
# 10000 original.report
$ gpdcreport -n 5000 original.report -o first_part.report
$ gpdcreport first_part.report -count
# 5000 first_part.report 
At this step we have a new report file that contains only the first 5000 models. You can run the following command to get the best model:

Code: Select all

$ gpdcreport first_part.report -best 1
# Layered model 4871: value=0.0300515
2
4.984615585360895551 1648.7744072668654098 251.53441286720578773 2000
0 3582.8480720604884482 838.858913229511586 2000
To get the profile for Vp, Vs, or Nu, pipe this to gpprofile and figue:

Code: Select all

$ gpdcreport first_part.report -best 1 | gpprofile -vs | figue -c
For the second part, if you are lucky enough, the best misfit of the second part is lower than in the first part. If so, you can run the same commands on the original.report, the best model will be located in the second part.

To get the second part as a separate report file, and if you can open the report inside dinver, you can show ground profiles for the original report file with a very high maximum misfit to select all models and specify the minimum and maximum index to select. In tool menu, choose "Save as report".

An alternative method is to select models in gpdcreport by a list of index, for instance:

Code: Select all

$ echo | awk '{for(i=5000; i<10000;i++) print i}' | gpdcreport -index original.report -o second_part.report
Best regards,

Marc
Post Reply