faststructure:--cv=5 --prior=simple --format=str --full --seed=100

[xiaoji...@gmail.com]

Now I am running thounsands of SNPs, I try set the "-CV=5". some minutes later,screen present:"Failed /n;Failed /n;Failed /n;Failed /n;Failed /n;Failed /n;"
   Anybody  have accounted the similar promble?
  or anybody knows  which rise that errors?

[Vikram Chhatre]

Can you show us complete log of your command and screen output?

V

--
You received this message because you are subscribed to the Google Groups "structure-software" group.
To unsubscribe from this group and stop receiving emails from it, send an email to structure-softw...@googlegroups.com.
To post to this group, send email to structure...@googlegroups.com.
Visit this group at http://groups.google.com/group/structure-software.
For more options, visit https://groups.google.com/d/optout.

[xiaoji...@gmail.com]

Thank you Vikram Chhatre, so quickly reponse my quzze.
  it doesnot produce the "*.log"  file, Now I attachment the printscreen;




PS:the next question is "--prior=logisitic" very very slow.
Marginal likelihood with initialization (1) = -0.4464247021
Marginal likelihood with initialization (2) = -0.3967253403
Marginal likelihood with initialization (3) = -0.4127808493
Marginal likelihood with initialization (4) = -0.4078836419
Marginal likelihood with initialization (5) = -0.3950017331
Iteration Marginal_Likelihood delta_Marginal_Likelihood Iteration_Time (secs)
0 -0.3950017331 -- 33966.505
10 -0.3949870790 0.0000146541 71.998
20 nan nan 167.471
Marginal Likelihood = nan
Total time = 34207.3447 seconds
Total iterations = 20


在 2015年9月17日星期四 UTC+8上午9:22:05，xiaoji...@gmail.com写道：

[sudapar...@gmail.com]

Hello,
I encountered the same problem today, were you able to understand why?
Best,
SR

[Stéphanie Arnoux]

Hi all,

I am also in this situation, I have obtain some plots but when changed the cv got the failed...

  File "structure.py", line 178, in <module>

    params['prior'], params['cv'])

  File "fastStructure.pyx", line 167, in fastStructure.infer_variational_parameters (fastStructure.c:4247)

    meandeviance = CV(G, psi, pi, cv, mintol)

  File "fastStructure.pyx", line 297, in fastStructure.CV (fastStructure.c:5886)

    Gmask = -1*np.ones((N,L), dtype='int8')

  File "/System/Library/Frameworks/Python.framework/Versions/2.7/Extras/lib/python/numpy/core/numeric.py", line 136, in ones

    def ones(shape, dtype=None, order='C'):

here is the Traceback...

Any suggestion?

Best,

Stéph

[Vikram Chhatre]

Stephanie,

I suggest emailing Anil Raj directly about this problem.

V

--

You received this message because you are subscribed to the Google Groups "structure-software" group.
To unsubscribe from this group and stop receiving emails from it, send an email to structure-softw...@googlegroups.com.
To post to this group, send email to structure...@googlegroups.com.

Visit this group at https://groups.google.com/group/structure-software.

[Stéphanie Arnoux]

Hi,

I indeed emailed him and never got a response back. I have kept my faststructure without cv and with colours that don't fit the K.

But I am really suspicious on how to justify my results... (I know from previous study that it looks quite ok but only for one of my species)

Best,

S