Error when perform correct_troposphere for hyp3 products #1057
Comments
|
I would suggest check the following:
|
Hi, your response is greatly appreciated. Anyway, I'll give it a try using GACOS. Thanks again. |
No, all methods should work as described. I was having the impression that you were trying to use While testing PyAPS/ERA5 on the Ridgecrest EQ example dataset, I had an error as well, which is now fixed by #1061. But the error I had is different from yours, could you test that PR to check if it fixes your issue as well? |
Hi, thanks for your prompt response. |
I attempted to correct the troposphere on the Ridgecrest dataset. Due to the fact that ERA5.h5 is only 10+kb, I believe there is an issue. Then, I tested the ERA5.h5 file provided with the Ridgecrest dataset and it worked well. |
|
Hmm... Could you check if your pyaps is installed and runs properly, by running the The output should look like below: (insar) yunjunz:~>$ python tools/PyAPS/tests/test_calc.py
------------------------------------------------
import pyaps3 from /Users/yunjunz/tools/PyAPS/src/pyaps3/__init__.py
------------------------------------------------
test tropospheric delay calculation from ERA5.
------------------------------------------------
read ISCE geometry files: hgt/los/lat/lon.rdr
calculate tropospheric delay from GRB files...
------------------------------------------------
INFO: INCIDENCE ANGLE AS AN ARRAY
INFO: AREA COVERAGE IN SNWE: (33.85, 30.05, 129.05, 132.45)
PROGRESS: READING GRIB FILE
INFO: USING PRESSURE LEVELS OF ERA-INT OR ERA-5 DATA
INFO: IMAGE DIMENSIONS: 15 LATITUDES AND 13 LONGITUDES
PROGRESS: INTERPOLATING FROM PRESSURE TO HEIGHT LEVELS
PROGRESS: COMPUTING DELAY FUNCTIONS
INFO: INCIDENCE ANGLE AS AN ARRAY
INFO: AREA COVERAGE IN SNWE: (33.85, 30.05, 129.05, 132.45)
PROGRESS: READING GRIB FILE
INFO: USING PRESSURE LEVELS OF ERA-INT OR ERA-5 DATA
INFO: IMAGE DIMENSIONS: 15 LATITUDES AND 13 LONGITUDES
PROGRESS: INTERPOLATING FROM PRESSURE TO HEIGHT LEVELS
PROGRESS: COMPUTING DELAY FUNCTIONS
PROGRESS: FINE INTERPOLATION OF HEIGHT LEVELS
PROGRESS: CREATE THE BILINEAR INTERPOLATION FUNCTION
PROGRESS: MAPPING THE DELAY
[============================================================] 0s / 0s
PROGRESS: FINE INTERPOLATION OF HEIGHT LEVELS
PROGRESS: CREATE THE BILINEAR INTERPOLATION FUNCTION
PROGRESS: MAPPING THE DELAY
[============================================================] 0s / 0s
------------------------------------------------
Passed tropospheric delay calculation from ERA5.
------------------------------------------------
|
After thoroughly examining the detailed error message, I believe it was not caused by mintpy or pyaps, but by an environmental configuration issue. I believe that numpy or scipy, or something else, is the cause of the issue, although it's not clear. So I re-configured the Conda environment and it now works well. Sorry for my own reasons, I mistakenly thought it was a bug. I believe we can close this comment this time. Thank you very much! |
|
No worries, I am glad to hear you locate the cause and fix the issue! |
Hi, it seems that the error is still present. I'm trying to print the message from some scripts. PyAPS/src/pyaps3/objects.py", line 177, in init Referring to the error message "ValueError: zero-size array to reduction operation maximum which has no identity", I think it was caused by the 'gph.max().round' function. Then I realized that the gph (array) was empty, and defined in era.py line 208: After following the output message step by step, I discovered that the coverage of lats(array) is different with the true latitude of the dataset. With the Ridgecrest dataset, the output message is as follows: We can see that, the output minlat (-1.199681364756279), maxlat(1.2003260023253484 ) are incorrect. These two arrays are calculated using the formula in PyAPS/src/pyaps3/objects.py, in lin 111: So, I think there may be some errors in calculating the min- and max-latitude. |
|
For errors in pyaps, please consider opening an issue on its repo instead. Nevertheless, I could successfully run Ridgecrest dataset with pyaps, so my guess is that your code is not properly installed. I would recommend re-installing them from scratch before digging into the code. Both PyAPS and MintPy have tests, use them to check your installation. |
|
Hmm, I reinstalled them yesterday following your guide on conda-envs. May I ask you to test the Ridgecrest dataset without the ERA5.h5 file under mintpy/inputs ? Thanks! |
I already did previously. If you could strictly follow the bug report template, and fill in all the details, it could help to replicate your issue. |
Problem description
Hi, wish you all the best.
I want to do the time series analysis on the iterferograms generated by Hyp3.
And I have pre-processed the tiffs accordingly.
When came to the "correct_troposphere" ,I encounter errors.
After checking the generated file "ERA5.h5," which was just 14.4 kilobytes, it became apparent that something was wrong.
I used info.py to view the information for "ERA5.h5", but it was full of zeros.
I used view.py to plot the "ERA5.h5", but it was empty.
I've tried using different sets of interferograms (also generated by Hyp3), and they all come out the same way.
Full script that generated the error
Full error message
System information
Any help is greatly appreciated.
The text was updated successfully, but these errors were encountered: