Step LiCSBAS13 problem

[bondangd45]

Dear Prof Yu,

I was installing the LiCSBAS on Ubuntu 16.04 by using Anaconda3. I have tried to follow the steps, but I could not install the gdal module by using: conda install gdal, directly. Then I tried to use that command line after I created the conda virtual environment. In here, I can use the gdal. But I also have to install the other modules including astropy, beautifulsoup4, matplotlib, h5py, numpy, requests, and statsmodels because I cannot use the modules directly from /home.
My steps is as follows:

1. Install the Anaconda in /home. Python version detected is Python 3.8.2
2. install the Anaconda environment by using: conda create --name myenv. Then I activate it.
3. In the Anaconda environment, I also install the modules required for LiCSBAS process
4. cd <your_LiCSBAS_path>
   git clone https://github.com/yumorishita/LiCSBAS.git
5. Make the directory which is used for the LiCSBAS processing
6. source <your_LiCSBAS_path>/LiCSBAS/bashrc_LiCSBAS.sh
7. setting the batch file and then run it by using: ./batch_LiCSBAS.sh

Then, I found the problem in the following text:
LiCSBAS13_sb_inv.py ver1.2 20200225 Y. Morishita
LiCSBAS13_sb_inv.py -d GEOCml1clip
In geographical coordinates

Size of image (w,l) : 202, 242

of all images : 3

of images to be used : 3

of all ifgs : 3

of ifgs to be used : 3

of removed ifgs : 0

Threshold of used unw : 3

Reference area (X/Y) : 194:195/193:194
Allowed memory size : 4000 MB
Number of patches : 1
Inversion algorism : LS
Gamma value : 0.0001

Process 242/242th line (1/1th patch)...
Reading 3 ifg's unw data...
48884/48884 points removed due to not enough ifg data...

Identifing gaps and counting n_gap...
Counting n_ifg_noloop...

Small Baseline inversion by LS...

Next, solve 0 points including nan point-by-point...

Elapsed time for 1th patch: 00h 00m 00s

Find stable reference point...
/home/bondan/anaconda3/envs/myenv/lib/python3.8/site-packages/numpy/lib/nanfunctions.py:1115: RuntimeWarning: All-NaN slice encountered
r, k = function_base._ureduce(a, func=_nanmedian, axis=axis, out=out,
/home/bondan/LiCSBAS/LiCSBAS/bin/LiCSBAS13_sb_inv.py:628: RuntimeWarning: All-NaN slice encountered
min_n_gap = np.nanmin(n_gap)
/home/bondan/LiCSBAS/LiCSBAS/bin/LiCSBAS13_sb_inv.py:634: RuntimeWarning: All-NaN slice encountered
min_rms = np.nanmin(rms_cum_wrt_med)
Traceback (most recent call last):
File "/home/bondan/LiCSBAS/LiCSBAS/bin/LiCSBAS13_sb_inv.py", line 753, in
sys.exit(main())
File "/home/bondan/LiCSBAS/LiCSBAS/bin/LiCSBAS13_sb_inv.py", line 636, in main
refy1s, refx1s = refy1s[0], refx1s[0] ## Only first index
IndexError: index 0 is out of bounds for axis 0 with size 0

[yumorishita]

The reason of the error is that the number of ifgs is too small (only 3) to carry out the time series analysis. What is your frame ID? Firstly the sufficient number of ifgs must be created and published. You can request processing of your frame ID of interest on LiCSAR portal (https://comet.nerc.ac.uk/COMET-LiCS-portal/, right column).

[bondangd45]

Okay Prof, I will try to use longer period. The frame ID is: 076D_09725_121107.
Actually I also this problem when I try to process in another frame ID, Prof. The frame ID is 119D_09370_100707.

[sazzath]

Are there any videos for installation for LiCSBAS ??? One video can solve 100 Problems !!

[yumorishita]

Sorry, there is no videos. The installation of LiCSBAS itself is very easy as is written in wiki, while setup of a python environment is complicated for users who are not familiar with python. It is difficult to show a general method to setup a python environment because every user has different situations. Plenty of information of python is available on the internet. Please google it.

[sazzath]

Is this linux base tools ?

[yumorishita]

Yes, but should work on Mac or Windows as well. LiCSBAS needs Python and bash. If you don't have a linux, I recommend using VMware to use linux on Windows.

[irmayolii2]

Hi Prof Yu
Why when I try to display timeseries with script "LiCSBAS_plot_ts.py -i TS_GEOCml1clip/cum_filt.h5 &"
it's error like this.

script to display timeseries
112A_09051_071616$ LiCSBAS_plot_ts.py -i TS_GEOCml1maskclip/cum_filt.h5 &

result
[1] 24466
LiCSBAS_2/LiCSBAS/112A_09051_071616$
LiCSBAS_plot_ts.py ver1.13.3 20210205 Y. Morishita
LiCSBAS_plot_ts.py -i TS_GEOCml1maskclip/cum_filt.h5

Reading TS_GEOCml1maskclip/cum_filt.h5
Traceback (most recent call last):
File "/media/geodesi/SSD_2TB/skripsi_irma/LiCSBAS_2/LiCSBAS/bin/LiCSBAS_plot_ts.py", line 361, in
refx1, refx2, refy1, refy2 = [int(s) for s in re.split('[:/]', refarea)]
File "/home/geodesi/miniconda3/envs/licsbas/lib/python3.8/re.py", line 231, in split
return _compile(pattern, flags).split(string, maxsplit)
TypeError: cannot use a string pattern on a bytes-like object

[yumorishita]

Although I cannot reproduce this error in my environment, I have modified the codes to avoid it because the error reports have been increasing.
7637465

Please update the LiCSBAS module and try again.

[IamTariqAshraf]

Have a nice day. I wanted to ask if i have used gamma for preprocessing like making interferograms and unwrapp files can i use LICSBAS for visulaization only?

[yumorishita]

Yes, you can use LiCSBAS_disp_img.py to display the gamma files.

[IamTariqAshraf]

Dear Sir, Yumorishita, Thank you so much for your kind response. Sorry to bother you again but can we do some thing more or we can only visualize the gamma processed data using LiCSBAS. I mean like plotting time series using LiCSBAS_plot_ts.py ? Since I am not sure if we can create .h5 file to be used in LiCSBAS_plot_ts.py script.

[yumorishita]

Technically it is possible to create .h5 from any format but I have no tool to convert gamma time series results to .h5. You should make it yourself if you want to use LiCSBAS_plot_ts.py.

[IamTariqAshraf]

Mr.Yumorishita , Thank you for your kind response