Project

General

Profile

Actions
Copy link

Bug #1990

closed

picking bug at FSU

Added by Scott Stagg over 12 years ago. Updated over 10 years ago.

Status:
Won't Fix or Won't Do
Priority:
Normal
Assignee:
-
Category:
-
Target version:
Appion/Leginon 2.2.0
Start date:
08/13/2012
Due date:
% Done:

0%

Estimated time:
Affected Version:
Appion/Leginon 2.2.0
Show in known bugs:
No
Workaround:

Description

Here is the bug I was describing:

Traceback (most recent call last):
  File "/panfs/storage.local/imb/stagg/software/rhel61/usr/local/bin/templateCorrelator.py", line 190, in <module>
    imgLoop.run()
  File "/panfs/storage.local/imb/stagg/software/myami_2_2env/lib/python2.6/site-packages/appionlib/appionLoop2.py", line 70, in run
    results = self.loopProcessImage(imgdata)
  File "/panfs/storage.local/imb/stagg/software/myami_2_2env/lib/python2.6/site-packages/appionlib/particleLoop2.py", line 104, in loopProcessImage
    self.peaktree = filterLoop.FilterLoop.loopProcessImage(self, imgdata)
  File "/panfs/storage.local/imb/stagg/software/myami_2_2env/lib/python2.6/site-packages/appionlib/filterLoop.py", line 87, in loopProcessImage
    peaktree = self.processImage(imgdata, self.filtarray)
  File "/panfs/storage.local/imb/stagg/software/rhel61/usr/local/bin/templateCorrelator.py", line 163, in processImage
    peaktree  = apPeaks.findPeaks(imgdata, ccmaplist, self.params)
  File "/panfs/storage.local/imb/stagg/software/myami_2_2env/lib/python2.6/site-packages/appionlib/apPeaks.py", line 50, in findPeaks
    maxpeaks, maxsizemult, msg, tmpldbid, mapdiam, bin=bin, peaktype=peaktype)
  File "/panfs/storage.local/imb/stagg/software/myami_2_2env/lib/python2.6/site-packages/appionlib/apPeaks.py", line 94, in findPeaksInMap
    varyThreshold(imgmap, thresh, maxsize)
  File "/panfs/storage.local/imb/stagg/software/myami_2_2env/lib/python2.6/site-packages/appionlib/apPeaks.py", line 330, in varyThreshold
    blobtree, percentcov = findBlobs(ccmap, thresh, maxsize=maxsize)
  File "/panfs/storage.local/imb/stagg/software/myami_2_2env/lib/python2.6/site-packages/appionlib/apPeaks.py", line 396, in findBlobs
    maxsize, minsize, maxmoment, elim, summary)
  File "/panfs/storage.local/imb/stagg/software/myami_2_2env/lib/python2.6/site-packages/pyami/imagefun.py", line 357, in find_blobs
    blobs = scipyblobs(image,tmpmask)
  File "/panfs/storage.local/imb/stagg/software/myami_2_2env/lib/python2.6/site-packages/pyami/imagefun.py", line 286, in scipyblobs
    stds = scipy.ndimage.standard_deviation(im,labels,range(1,n+1))
  File "/usr/lib64/python2.6/site-packages/scipy/ndimage/measurements.py", line 649, in standard_deviation
    return numpy.sqrt(variance(input, labels, index))
  File "/usr/lib64/python2.6/site-packages/scipy/ndimage/measurements.py", line 596, in variance
    count, sum, sum_c_sq = _stats(input, labels, index, centered=True)
  File "/usr/lib64/python2.6/site-packages/scipy/ndimage/measurements.py", line 434, in _stats
    sums_c = _sum_centered(labels.ravel())
  File "/usr/lib64/python2.6/site-packages/scipy/ndimage/measurements.py", line 408, in _sum_centered
    centered_input = input - means[labels]
ValueError: operands could not be broadcast together with shapes (512,512) (262144) 
[sstagg@kriosdb ~]$

Actions
Copy link
 #1

Updated by Scott Stagg about 12 years ago

I found a work around for this. It seems that with the setup at FSU scipy.ndimage.standard_deviation(im,labels,range(1,n+1)) is not behaving the same as it does at Scripps. I think this must be a problem with the particular numpy, scipy, python combination that we are using at FSU. I got around the error by replacing the scipy.ndimage.standard_deviation(im,labels,range(1,n+1)) line in scipyblobs function with:

stds = numpy.zeros(n)

This is a less than desirable workaround. Anchi and Jim, do you have any suggestions about what else I could do? BTW, I am using python 2.6.6, scipy 0.9.0, and numpy 1.6.1

Actions
Copy link
 #2

Updated by Anchi Cheng about 12 years ago

How about upgrade to scipy 0.10? See #1401

Actions
Copy link
 #3

Updated by Anchi Cheng about 12 years ago

Option 2 (works on my Mac that suffers the same problem even after I upgraded to 0.10)

  1. Go to you scipy installation in python site-package and edit ndimage/measurements.py
  2. Find the function _sum_centered and modify the following:

Change

        centered_input = input - means[labels]

To
        label_means = numpy.reshape(means[labels],input.shape)
        centered_input = input - label_means

Actions
Copy link
 #4

Updated by Amber Herold over 10 years ago

  • Assignee deleted (Amber Herold)
Actions
Copy link
 #5

Updated by Amber Herold over 10 years ago

  • Status changed from New to Won't Fix or Won't Do
Actions
Copy link

Also available in: Atom PDF