26 Nov
2013
26 Nov
'13
11:37 a.m.
Xiao,
You are creating an array of :
100 (samples) * 86400 (seconds) * 22 (stations) = 190.080.000 samples.
On my machine, a float (1.0 e.g.) is 24 bytes, so your array will be :
4.350 MB (=4GB), which is too big, even on my machine.
You should downsample your data (cc_sampling_rate) to something closer
to 2 times the High frequency, to save memory: 10 Hz would be enough.
Let me know,
Thomas
Le 26/11/2013 11:27, Wang, Xiao a écrit :
Hi Thom,
I got 22 stations, miniseed sample rate is 100Hz, cc_sampling_rate is 100.0, mov_stack is
5, days between startdate and enddate is one month, and ZZ correlation is enabled only.
The filter parameters are: Low=0.1, Mwcs low=0.2, Mwcs high=0.6, High=0.7, Rms
threshold=0.0, Mwcs wlen=50.0 and Mwcs step=10.0
Xiao
-----Original Message-----
From: msnoise-bounces(a)mailman-as.oma.be [mailto:msnoise-bounces@mailman-as.oma.be] On
Behalf Of Thomas Lecocq
Sent: 25 November 2013 18:06
To: msnoise(a)mailman-as.oma.be
Subject: Re: [MSNoise] MemoryError
Dear Xiao,
How many stations pairs are you processing simultenously ? 1 day of data ? 20Hz ? Please
provide more details, it'll help !
Cheers
Thom
Le 25/11/2013 18:39, Wang, Xiao a écrit :
When I ran python s03compute_cc.py (msnoise
version1.2), I encountered memoryerror. Any idea?
The following is output:
Traceback (most recent call last):
File "s03compute_cc.py", line 191, in <module>
tramef_Z = np.zeros((len(stations), len(TimeVec))) MemoryError
Thanks,
Xiao
_______________________________________________
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
_______________________________________________
MSNoise mailing list
MSNoise(a)mailman-as.oma.be
http://mailman-as.oma.be/mailman/listinfo/msnoise