2

I am trying to run parallel code in python utilizing ipyparallel package. 

import ipyparallel as ipp
# create client & view
rc = ipp.Client()
dv = rc[:]
v = rc.load_balanced_view()

def my(i):    
     import numpy as np
     aa[:,:,i]=np.dot(uanom_L[:,:,i-I_a:i+I_a+1],lcoz_a)
     bb[:,:,i]=np.dot(uanom_L[:,:,i-I_b:i+I_b+1],lcoz_b) 
     print(i)
     return aa,bb       

aaa=v.map(my,np.arange(I_a,c-I_a))
aaa.get()

Yet, I got the following error, 

[3:apply]: No traceback available
[2:apply]: No traceback available
[1:apply]: No traceback available
[0:apply]: No traceback available

8773 more exceptions ...

Any ideas Thanks

kmario23
  • 57,311
  • 13
  • 161
  • 150
Kernel
  • 591
  • 12
  • 23
  • You might check the logs in the controller to see if there was something about failing to produce the traceback. What are the versions of: ipython, ipykernel, ipyparallel. Also, the following information is missing from the example: I_a, c, I_b, lcoz_a, lcoz_b, uanom_L – minrk Oct 03 '16 at 09:56
  • Thanks a lot, I am wondering how can I check from the controller?. Here is the version: ipykernel-4.3.1-py27_0, ipyparallel-5.2.0-py27_0, ipython-4.1.2-py27_1. Regarding to the missing information--I_a, c, I_b, Icoz_a, Icoz_b, uanom_L--- I can send you npz file containing all those variables. Thanks – Kernel Oct 05 '16 at 00:33

0 Answers0