[MSNoise] problem in compute_dtt (master version with ACC)

Thomas Lecocq thomas.lecocq at oma.be
Fri Apr 7 07:58:53 UTC 2017 

Cool,

there was no real reason for hardcoding this value. I've ported the 
change to github:master already.

Cheers

Tom


Le 07/04/2017 à 09:53, Christos Evangelidis a écrit :
> I have deleted in line 1500 the "xtol=1e-20" and it works.
>
> Christos
>
> On Fri, Apr 7, 2017 at 10:46 AM, Thomas Lecocq <thomas.lecocq at oma.be> wrote:
>
>> It looks like the xtol is still there... although it indeed looks like
>> another error.
>>
>>
>>
>> Le 07/04/2017 à 09:42, Christos Evangelidis a écrit :
>>
>>> But now it stops after a while with this:
>>>
>>> raceback (most recent call last):
>>>     File "/home/noise/anaconda2/bin/msnoise", line 11, in <module>
>>>       load_entry_point('msnoise==1.5a0', 'console_scripts', 'msnoise')()
>>>     File
>>> "/home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>> scripts/msnoise.py",
>>> line 708, in run
>>>       cli(obj={})
>>>     File "/home/noise/anaconda2/lib/python2.7/site-packages/click/cor
>>> e.py",
>>> line 722, in __call__
>>>       return self.main(*args, **kwargs)
>>>     File "/home/noise/anaconda2/lib/python2.7/site-packages/click/cor
>>> e.py",
>>> line 697, in main
>>>       rv = self.invoke(ctx)
>>>     File "/home/noise/anaconda2/lib/python2.7/site-packages/click/cor
>>> e.py",
>>> line 1066, in invoke
>>>       return _process_result(sub_ctx.command.invoke(sub_ctx))
>>>     File "/home/noise/anaconda2/lib/python2.7/site-packages/click/cor
>>> e.py",
>>> line 895, in invoke
>>>       return ctx.invoke(self.callback, **ctx.params)
>>>     File "/home/noise/anaconda2/lib/python2.7/site-packages/click/cor
>>> e.py",
>>> line 535, in invoke
>>>       return callback(*args, **kwargs)
>>>     File
>>> "/home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>> scripts/msnoise.py",
>>> line 421, in compute_dtt
>>>       main(interval)
>>>     File
>>> "/home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>> s06compute_dtt.py",
>>> line 359, in main
>>>       intercept_origin=True)
>>>     File "/home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>> api.py",
>>> line 1500, in linear_regression
>>>       xtol=1e-20)
>>>     File
>>> "/home/noise/anaconda2/lib/python2.7/site-packages/scipy/opt
>>> imize/minpack.py",
>>> line 740, in curve_fit
>>>       raise RuntimeError("Optimal parameters not found: " + errmsg)
>>> RuntimeError: Optimal parameters not found: gtol=0.000000 is too small,
>>> func(x) is orthogonal to the columns of
>>>     the Jacobian to machine precision.
>>>
>>>
>>> On Fri, Apr 7, 2017 at 10:41 AM, Christos Evangelidis <cevan at noa.gr>
>>> wrote:
>>>
>>> Thank you!
>>>> It works!
>>>>
>>>> Christos
>>>>
>>>> On Fri, Apr 7, 2017 at 10:34 AM, Thomas Lecocq <thomas.lecocq at oma.be>
>>>> wrote:
>>>>
>>>> Christos
>>>>> could you edit the api:
>>>>>
>>>>> /home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>>>> api.py",
>>>>> line 1508, in linear_regression
>>>>>
>>>>> and remove the parameter xtol=1e-20, so scipy would use the default 1e-8
>>>>> or so ; reset your DTT jobs and relaunch ?
>>>>>
>>>>> Thanks
>>>>>
>>>>> Thomas
>>>>>
>>>>>
>>>>>
>>>>> Le 07/04/2017 à 09:28, Christos Evangelidis a écrit :
>>>>>
>>>>> Hi Thomas,
>>>>>> Yes I am using the latest master version.
>>>>>>
>>>>>> Christos
>>>>>>
>>>>>>
>>>>>> On Fri, Apr 7, 2017 at 10:22 AM, Thomas Lecocq <thomas.lecocq at oma.be>
>>>>>> wrote:
>>>>>>
>>>>>> Hi Christos,
>>>>>>
>>>>>>> you're using the latest master version ?
>>>>>>>
>>>>>>> It looks like the linear regression complains ; will have to look into
>>>>>>> it
>>>>>>> before 1.5 ... :(
>>>>>>>
>>>>>>> Thomas
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Le 06/04/2017 à 12:42, Christos Evangelidis a écrit :
>>>>>>>
>>>>>>> Hi all,
>>>>>>>
>>>>>>>> I have the following error trying to calculate the DTT with the
>>>>>>>> command
>>>>>>>> (master version)
>>>>>>>>
>>>>>>>> msnoise compute_dtt -i 999
>>>>>>>>
>>>>>>>> for 2 stations with autocorrelations. It seems that the script stacks
>>>>>>>> in a
>>>>>>>> some days (eg. 2011-02-27).
>>>>>>>>
>>>>>>>> Any idea?
>>>>>>>>
>>>>>>>> Cheers,
>>>>>>>>
>>>>>>>> Christos
>>>>>>>>
>>>>>>>> -----------------------------------------------
>>>>>>>> 2017-04-06 10:37:45 [INFO] *** Starting: Compute DT/T ***
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] Found 2229 updated days
>>>>>>>> 2017-04-06 10:37:45 [INFO] Loading mov=5 days for filter=02
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] Processing 2011-02-03 - 02 - 05 mov
>>>>>>>> /home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>>>>>>> s06compute_dtt.py:140:
>>>>>>>> RuntimeWarning: invalid value encountered in double_scalars
>>>>>>>>       wavg = (d * w).sum() / w.sum()
>>>>>>>> /home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>>>>>>> s06compute_dtt.py:142:
>>>>>>>> RuntimeWarning: invalid value encountered in double_scalars
>>>>>>>>       wstd = np.sqrt(np.sum(w * (d - wavg) ** 2) / ((N - 1) *
>>>>>>>> np.sum(w)
>>>>>>>> / N))
>>>>>>>> /home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>>>>>>> s06compute_dtt.py:340:
>>>>>>>> RuntimeWarning: invalid value encountered in less_equal
>>>>>>>>       errindex = np.where(errArray[i] <= maxErr)[0]
>>>>>>>> /home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>>>>>>> s06compute_dtt.py:341:
>>>>>>>> RuntimeWarning: invalid value encountered in less_equal
>>>>>>>>       dtindex = np.where(np.abs(dtArray[i]) <= maxDt)[0]
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] 2011-02-03: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] Processing 2011-02-04 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] 2011-02-04: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] Processing 2011-02-05 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] 2011-02-05: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] Processing 2011-02-06 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] 2011-02-06: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] Processing 2011-02-07 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] 2011-02-07: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] Processing 2011-02-08 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] 2011-02-08: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:45 [DEBUG] Processing 2011-02-09 - 02 - 05 mov
>>>>>>>> /home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>>>>>>> s06compute_dtt.py:142:
>>>>>>>> RuntimeWarning: divide by zero encountered in double_scalars
>>>>>>>>       wstd = np.sqrt(np.sum(w * (d - wavg) ** 2) / ((N - 1) *
>>>>>>>> np.sum(w)
>>>>>>>> / N))
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-09: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-10 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-10: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-11 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-11: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-13 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-13: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-14 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-14: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-15 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-15: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-16 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-16: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-17 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-17: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-18 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-18: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-19 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-19: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-20 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-20: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-21 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-21: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-22 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-22: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-23 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-23: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-24 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-24: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-25 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-25: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-26 - 02 - 05 mov
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] 2011-02-26: exporting: 4 pairs
>>>>>>>> 2017-04-06 10:37:46 [DEBUG] Processing 2011-02-27 - 02 - 05 mov
>>>>>>>> Traceback (most recent call last):
>>>>>>>>       File "/home/noise/anaconda2/bin/msnoise", line 11, in <module>
>>>>>>>>         load_entry_point('msnoise==1.5a0', 'console_scripts',
>>>>>>>> 'msnoise')()
>>>>>>>>       File
>>>>>>>> "/home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>>>>>>> scripts/msnoise.py",
>>>>>>>> line 708, in run
>>>>>>>>         cli(obj={})
>>>>>>>>       File "/home/noise/anaconda2/lib/pyt
>>>>>>>> hon2.7/site-packages/click/cor
>>>>>>>> e.py",
>>>>>>>> line 722, in __call__
>>>>>>>>         return self.main(*args, **kwargs)
>>>>>>>>       File "/home/noise/anaconda2/lib/pyt
>>>>>>>> hon2.7/site-packages/click/cor
>>>>>>>> e.py",
>>>>>>>> line 697, in main
>>>>>>>>         rv = self.invoke(ctx)
>>>>>>>>       File "/home/noise/anaconda2/lib/pyt
>>>>>>>> hon2.7/site-packages/click/cor
>>>>>>>> e.py",
>>>>>>>> line 1066, in invoke
>>>>>>>>         return _process_result(sub_ctx.command.invoke(sub_ctx))
>>>>>>>>       File "/home/noise/anaconda2/lib/pyt
>>>>>>>> hon2.7/site-packages/click/cor
>>>>>>>> e.py",
>>>>>>>> line 895, in invoke
>>>>>>>>         return ctx.invoke(self.callback, **ctx.params)
>>>>>>>>       File "/home/noise/anaconda2/lib/pyt
>>>>>>>> hon2.7/site-packages/click/cor
>>>>>>>> e.py",
>>>>>>>> line 535, in invoke
>>>>>>>>         return callback(*args, **kwargs)
>>>>>>>>       File
>>>>>>>> "/home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>>>>>>> scripts/msnoise.py",
>>>>>>>> line 421, in compute_dtt
>>>>>>>>         main(interval)
>>>>>>>>       File
>>>>>>>> "/home/noise/anaconda2/lib/python2.7/site-packages/msnoise/
>>>>>>>> s06compute_dtt.py",
>>>>>>>> line 356, in main
>>>>>>>>         intercept_origin=False)
>>>>>>>>       File "/home/noise/anaconda2/lib/pyt
>>>>>>>> hon2.7/site-packages/msnoise/
>>>>>>>> api.py",
>>>>>>>> line 1508, in linear_regression
>>>>>>>>         xtol=1e-20)
>>>>>>>>       File
>>>>>>>> "/home/noise/anaconda2/lib/python2.7/site-packages/scipy/opt
>>>>>>>> imize/minpack.py",
>>>>>>>> line 740, in curve_fit
>>>>>>>>         raise RuntimeError("Optimal parameters not found: " + errmsg)
>>>>>>>> RuntimeError: Optimal parameters not found: xtol=0.000000 is too
>>>>>>>> small, no
>>>>>>>> further improvement in the approximate
>>>>>>>>       solution is possible.
>>>>>>>> _______________________________________________
>>>>>>>> MSNoise mailing list
>>>>>>>> MSNoise at mailman-as.oma.be
>>>>>>>> http://mailman-as.oma.be/mailman/listinfo/msnoise
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>>
>>>>>>> MSNoise mailing list
>>>>>>> MSNoise at mailman-as.oma.be
>>>>>>> http://mailman-as.oma.be/mailman/listinfo/msnoise
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>> _______________________________________________
>>>>> MSNoise mailing list
>>>>> MSNoise at mailman-as.oma.be
>>>>> http://mailman-as.oma.be/mailman/listinfo/msnoise
>>>>>
>>>>>
>>>> --
>>>> ---------------------------------------------------------------
>>>> Dr. Christos Evangelidis
>>>> Associate Researcher - Seismologist
>>>> Hellenic Broadband Seismic Network
>>>> INSTITUTE OF GEODYNAMICS
>>>> NATIONAL OBSERVATORY OF ATHENS
>>>> Tel:+30 210 3490043 <+30%2021%200349%200043>
>>>> Email: cevan at noa.gr
>>>> Www: http://members.noa.gr/cevan/en/
>>>> ---------------------------------------------------------------
>>>>
>>>>
>>>
>> _______________________________________________
>> MSNoise mailing list
>> MSNoise at mailman-as.oma.be
>> http://mailman-as.oma.be/mailman/listinfo/msnoise
>>
>
>

More information about the MSNoise mailing list