13 May
2014
13 May
'14
2:14 a.m.
Thomas, thanks for the reply. I did find some empty DTT days, and fixed
that. Now, I get a different error:
/usr/lib64/python2.7/site-packages/pandas/core/common.py:195:
DeprecationWarning: numpy boolean negative (the unary `-` operator) is
deprecated, use the bitwise_xor (the `^` operator) or the logical_xor
function instead.
return -res
Traceback (most recent call last):
File "s07plot_dtt.py", line 130, in <module>
plt.fill_between(ALL.index,ALL[dttname]-ALL[errname],ALL[dttname]+ALL[errname],lw=1,color='red',zorder=-1,alpha=0.3)
File "/usr/lib64/python2.7/site-packages/matplotlib/pyplot.py", line
2757, in fill_between
interpolate=interpolate, **kwargs)
File "/usr/lib64/python2.7/site-packages/matplotlib/axes.py", line 6988,
in fill_between
x = ma.masked_invalid(self.convert_xunits(x))
File "/usr/lib64/python2.7/site-packages/numpy/ma/core.py", line 2244, in
masked_invalid
condition = ~(np.isfinite(a))
TypeError: ufunc 'isfinite' not supported for the input types, and the
inputs could not be safely coerced to any supported types according to the
casting rule ''safe''
But if I comment out the line that caused this, the process works, and
attached is the output. All is quiet under Auckland, I am happy to report.
Cheers,
Kasper
On 12 May 2014 19:02, Thomas Lecocq <thomas.lecocq(a)oma.be> wrote:
Kasper,
When you changed the mov_stack value, did you re-run MSNoise "from
scratch". The thing is, if jobs have already been done, say, for January
1 -> January 10, then you modify the mov_stack, and run s04_stack.py a
few days after, Jan1->10 will not be stacked for the 30 day mov_stack.
you have, at least, to run s04stack.py with a "--interval X" (X > 1,
your case probably 20 or 30 max), so the CC jobs in the last 30 days
will be re-considered for stacking.
As of the plot problem, I've noticed a slight bug in the get_stations()
in the database_tools, but that should only affect the results iif the
stations were input manually in the database and thus not necessarily in
alphabetical order:
(bugfix:
https://github.com/ROBelgium/MSNoise/commit/0dca3cfe101c56f3ed8f6e537412857…
).
In your case, it looks like the 1-day data is not loaded, could you
confirm the data is present in the folders/files archive ? You could
also print the "day" variable, to know which file is "empty" and
causes
the bug. Normally, empty dtt files should not exist...
Let me know,
Tom
Le 09/05/2014 02:18, Kasper van Wijk a écrit :
Dear All,
The whole cron.sh was humming a long nicely on our MSNoise network,
until I
changed from stacking 1,2 and 5 days (initially)
to 1,2,3,4,5,30 days.
Now,
all steps seem to run UNTIL the plot command.
Here is the error (after
putting a print statement on the cause of the problem, df):
(I run as sudo, because root is running the cron.sh overnight:)
[kasper@localhost MSNoise-master]$ sudo python s07plot_dtt.py
loading 1 days
Empty DataFrame
Columns: [A, EA, EM, EM0, M, M0, Pairs]
Index: []
Empty DataFrame
Columns: [A, EA, EM, EM0, M, M0, Pairs]
Index: []
Traceback (most recent call last):
File "s07plot_dtt.py", line 102, in <module>
alldf = alldf.append(df)
File "/usr/lib64/python2.7/site-packages/pandas/core/frame.py", line
4266, in append
verify_integrity=verify_integrity)
File "/usr/lib64/python2.7/site-packages/pandas/tools/merge.py", line
883, in concat
return op.get_result()
File "/usr/lib64/python2.7/site-packages/pandas/tools/merge.py", line
964, in get_result
new_data = self._get_concatenated_data()
File "/usr/lib64/python2.7/site-packages/pandas/tools/merge.py", line
1007, in _get_concatenated_data
new_data[item] = self._concat_single_item(rdata, item)
File "/usr/lib64/python2.7/site-packages/pandas/tools/merge.py", line
1094, in _concat_single_item
return com._concat_compat(to_concat, axis=self.axis - 1)
File "/usr/lib64/python2.7/site-packages/pandas/core/common.py", line
1191, in _concat_compat
axis=axis)
ValueError: need at least one array to concatenate
_______________________________________________
MSNoise mailing list
MSNoise(a)mailman-as.oma.be
http://mailman-as.oma.be/mailman/listinfo/msnoise
_______________________________________________
MSNoise mailing list
MSNoise(a)mailman-as.oma.be
http://mailman-as.oma.be/mailman/listinfo/msnoise
