I’m re-writing StaMPS to Python/ Numpy. I have ported some StaMPS code to Python and tested the results. So far it seems good.
But I still need help with StaMPS code. There are some functions and variables that aren’t very well understandable and I want to improve readability in my code.
Can anyone here help me with that?
Any idea how fast that is likely to perform? Isn’t it difficult to parallelize Python?
I think the idea is great because it would make StaMPS available to non Linux/Matlab users. Maybe you find some people here: How to prepare Sentinel-1 images stack for PSI/SBAS in SNAP 5
It is surely a good idea to start a new topic on this but maybe you can leave a quick message in the topic mentioned above so that anyone interested notices it.
At this moment I’m just re-writing Matlab code to Python/ Numpy. This gives more readable and maintainable codebase. Also after that we don’t need Matlab licence. After that I think it is possible to make it run faster and use parallelization.
I will write to that other topic.
How is it going about re-writing matlab code to python? I think it is a good chance to update some parameters in StaMPS to process sentinel-1 data, I have already read the main code of PS processing, maybe I can provide some help.
Is there any news on this topic? Had you tried to contact Prof. Hooper on this subject? I guess it might be a good idea.
If you have it already done, I can try it on my areas to help you in the verification/validation of your new code.
Let me know
How did you get over it?
Here is the GitHub repo with Vant’s python translation of StAMPS (I hope he does not mind, the license is open):
As stated on the github! Please take this into consideration:
This isn’t complete StaMPS rewrite! This only contains steps that are responsible finding persistent scatterers. There are missing phase unwrapping for complete and standalone work.
The rewrite was done from StaMPS version 3.3b1.