preprocessing

[Emmanuel Hanert]

This topic deals with all the issues related to the preprocessing step.

[Ruth Patterson]

I did a system upgrade and I upgraded slim as well. I made a few other changes that I didn’t think would affect slim. However now slim doesn’t run. I deleted slim and reinstalled it according to the wiki instructions, and got some errors.

$ make -j4
CMakeFiles/dg.dir/flags.make:11: *** missing separator. Stop.
CMakeFiles/Makefile2:99: recipe for target ‘CMakeFiles/dg.dir/all’ failed
make[1]: *** [CMakeFiles/dg.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs….

the command kept processing for a while, but finished with

Makefile:94: recipe for target ‘all’ failed
make: *** [all] Error 2

When testing slim I got this error:

$ slim pre.py
Traceback (most recent call last):
File « pre.py », line 1, in <module>
import slimPre
File « /home/mlelab/slim/build/slimPre.py », line 97, in <module>
import dgpy
File « /home/mlelab/slim/build/dgpy/__init__.py », line 28, in <module>
from dgpy.mesh import *
ModuleNotFoundError: No module named ‘dgpy.mesh’

Is this a slim error or is there something unrelated to slim that’s causing it to fail. I’m using Linux (Ubuntu).

[jonathan]

Hi Ruth,

the problem comes from your python/numpy setup.

CMake detected python3.6 and found numpy in /home/mlelab/.local/lib/python3.6/site-packages/numpy/ but it seems that this combination is broken on your computer.

I do not really understand why you have a manually-installed numpy, you can install it from official ubuntu packages.

You have to fix your numpy setup. To check if it works (without using slim cmake), you can simply type this command in a terminal :

python3.6 -c « import numpy »

and it should not print any error.

(replace python3.6 by the python version you use if you are not using python3.6)

Jonathan

[Ruth Patterson]

Thanks Jonathan, this is fixed now.

[Ruth Patterson]

When I run SLIM on a windows10 subsystem for Linux the pre-processing fails to retrieve topex tides. This script works fine on Linux, and it also runs in bash if the data_1/0 folder is pre populated with a tide.nc file. Here is the error:

…
Info : bath smoothing iteration 40 : 0.00937009 (target 0.01)
Info : Projection system for tags:
Info :
Info : Projection used:
Info : mesh proj: +proj=utm +ellps=WGS84 +zone=54 +south
Info : data proj: +proj=latlong +ellps=WGS84
Info : Projection system for tags:
Info : arafurasea
Info : Projection used:
Info : mesh proj: +proj=utm +ellps=WGS84 +zone=54 +south
Info : data proj: +proj=latlong +ellps=WGS84
downloading h_tpxo7.2.nc of size 427M at /home/ruth/.cache/slim//slim_data/tides/h_tpxo7.2.nc
It can last a few minutes (or more, depending on your connection)
Traceback (most recent call last):
File « pre.py », line 43, in <module>
(ssh_tide, ux_tide, uy_tide) = slimPre.tpxo_tide(region_open, time_tpxo, data_file_name=data_dir+’open_bnd.nc’, export_as_transport = False)
File « /mnt/c/Users/rpatterson1/slim/build/slimPre.py », line 650, in tpxo_tide
h_file_name = slim_private.dgftp.get( (« /slim_data/tides/%s » % h_file_name) , checkdate = False)
File « /mnt/c/Users/rpatterson1/slim/build/dgpy/scripts/dgftp.py », line 86, in get
http://ftp.retrbinary(« RETR  » + filename, open(dst+ »_tmp », ‘wb’).write)
File « /usr/lib/python3.5/ftplib.py », line 441, in retrbinary
with self.transfercmd(cmd, rest) as conn:
File « /usr/lib/python3.5/ftplib.py », line 398, in transfercmd
return self.ntransfercmd(cmd, rest)[0]
File « /usr/lib/python3.5/ftplib.py », line 382, in ntransfercmd
resp = self.sendcmd(cmd)
File « /usr/lib/python3.5/ftplib.py », line 272, in sendcmd
return self.getresp()
File « /usr/lib/python3.5/ftplib.py », line 235, in getresp
resp = self.getmultiline()
File « /usr/lib/python3.5/ftplib.py », line 221, in getmultiline
line = self.getline()
File « /usr/lib/python3.5/ftplib.py », line 203, in getline
line = self.file.readline(self.maxline + 1)
File « /usr/lib/python3.5/socket.py », line 575, in readinto
return self._sock.recv_into(b)
socket.timeout: timed out

[Valentin Vallaeys]

Dear Ruth,

The problem seems related either to your internet connection, as the last line says « timed out » when trying to download the files, or to the directory path to store the files.

You may either try with a different internet connection, or copy both TPXO files from the directory in Linux (/home/ruth/.cache/slim/slim_data/tides/) to the equivalent directory on Windows (same path).

Best,
Valentin

[Ruth Patterson]

Thanks!

[Author]

Posts