search

kzwkt / wnd-charm

Automatically exported from code.google.com/p/wnd-charm
0 stars 0 forks source link

cannot use -f switch with classify command #18

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago 
What steps will reproduce the problem?
1.Navigate to  [eckleyd@lgopt14 CompareHVscans]$ pwd
/home/eckleyd/ICAWormImaging/scans/CompareHVscans

2.run [eckleyd@lgopt14 CompareHVscans]$ wndchrm classify -f.03  
/home/eckleyd/ICAWormImaging/Working/NewNotRotated/NewNR.fit Worm1.fit 

3.[eckleyd@lgopt14 CompareHVscans]$ wndchrm classify 
/home/eckleyd/ICAWormImaging/Working/NewNotRotated/NewNR.fit Worm1.fit 

What is the expected output? line 3. generates a set of marginal probabilities.
 What do you see instead? line 2. generates segmentation fault

What version of the product are you using?
wndchrm 1.27-101
Laboratory of Genetics/NIA/NIH 

 On what operating system?
Mac OS X, version 10.5.8

Please provide any additional information below.

Original issue reported on code.google.com by dmarkeck...@gmail.com on 8 Mar 2011 at 10:27

GoogleCodeExporter commented 9 years ago 
Mark, try using the latest and greatest version of wndchrm in your testing. I 
think I saw you already compile it. Just specify the full path to that newly 
compiled wndchrm executable and re-run your experiment, and report back here 
what happened.

Original comment by christop...@gmail.com on 8 Mar 2011 at 10:44

GoogleCodeExporter commented 9 years ago 
sure, great. tried and failed to work, same output:
[eckleyd@lgopt15 wndchrm-1.30.124_138]$ ./wndchrm classify -f.03 
/lgopt14/eckleyd/ICAWormImaging/Working/NewNotRotated/NewNR.fit 
/lgopt14/eckleyd/ICAWormImaging/scans/CompareHVscans/Worm1.fit 
Processing training set 
'/lgopt14/eckleyd/ICAWormImaging/Working/NewNotRotated/NewNR.fit'.
----------
Summary of '/lgopt14/eckleyd/ICAWormImaging/Working/NewNotRotated/NewNR.fit' 
(994 samples total, 1 samples per image):
'Class label' (interpreted value) number of samples.
'BackgroundNR'  (0) 526
'BodyNR'    (0) 223
'HeadNR'    (0) 135
'TailNR'    (0) 110
----------
Processing test set 
'/lgopt14/eckleyd/ICAWormImaging/scans/CompareHVscans/Worm1.fit'.
----------
Summary of '/lgopt14/eckleyd/ICAWormImaging/scans/CompareHVscans/Worm1.fit' (34 
samples total, 1 samples per image):
34 unknown samples. Suitable as a test/classification set only.
----------
Max balanced training: 110
samples per image=1, training images: 110, testing images 0
image   norm. fact. p(BackgroundNR) p(BodyNR)   p(HeadNR)   p(TailNR)   act. 
class   pred. class
Segmentation fault
[eckleyd@lgopt15 wndchrm-1.30.124_138]$

Original comment by dmarkeck...@gmail.com on 9 Mar 2011 at 3:36

GoogleCodeExporter commented 9 years ago 
[deleted comment]
GoogleCodeExporter commented 9 years ago 
problem appears to be using the -f switch-

[eckleyd@lgopt15 wndchrm-1.30.124_138]$ ./wndchrm classify 
/lgopt14/eckleyd/ICAWormImaging/Working/NewNotRotated/NewNR.fit 
/lgopt14/eckleyd/ICAWormImaging/scans/CompareHVscans/Worm1.fit 
Processing training set 
'/lgopt14/eckleyd/ICAWormImaging/Working/NewNotRotated/NewNR.fit'.
----------
Summary of '/lgopt14/eckleyd/ICAWormImaging/Working/NewNotRotated/NewNR.fit' 
(994 samples total, 1 samples per image):
'Class label' (interpreted value) number of samples.
'BackgroundNR'  (0) 526
'BodyNR'    (0) 223
'HeadNR'    (0) 135
'TailNR'    (0) 110
----------
Processing test set 
'/lgopt14/eckleyd/ICAWormImaging/scans/CompareHVscans/Worm1.fit'.
----------
Summary of '/lgopt14/eckleyd/ICAWormImaging/scans/CompareHVscans/Worm1.fit' (34 
samples total, 1 samples per image):
34 unknown samples. Suitable as a test/classification set only.
----------
Max balanced training: 110
samples per image=1, training images: 110, testing images 0
image   norm. fact. p(BackgroundNR) p(BodyNR)   p(HeadNR)   p(TailNR)   act. 
class   pred. class
./StraightWorm1/StraightenedWormtif_scan_01.tif 8.34e-10    0.000   0.760   0.016   0.223
    UNKNOWN BodyNR
./StraightWorm1/StraightenedWormtif_scan_02.tif 1.87e-10    0.000   0.739   0.051   0.210
    UNKNOWN BodyNR
./StraightWorm1/StraightenedWormtif_scan_03.tif 2.43e-07    0.000   0.131   0.037   0.832
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_04.tif 1.79e-08    0.000   0.001   0.010   0.989
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_05.tif 3.21e-06    0.000   0.000   0.020   0.980
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_06.tif 2.81e-07    0.000   0.001   0.017   0.983
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_07.tif 2.61e-08    0.000   0.005   0.006   0.990
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_08.tif 6.33e-07    0.000   0.041   0.010   0.948
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_09.tif 3.09e-06    0.000   0.000   0.005   0.995
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_10.tif 7.18e-07    0.000   0.026   0.334   0.639
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_11.tif 1.98e-05    0.000   0.000   0.035   0.965
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_12.tif 1.64e-08    0.000   0.002   0.011   0.987
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_13.tif 4.04e-06    0.000   0.000   0.002   0.998
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_14.tif 0.000304    0.000   0.000   0.003   0.997
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_15.tif 7.5e-12 0.000   1.000   0.000   0.000   
UNKNOWN BodyNR
./StraightWorm1/StraightenedWormtif_scan_16.tif 4.66e-10    0.000   0.891   0.007   0.102
    UNKNOWN BodyNR
./StraightWorm1/StraightenedWormtif_scan_17.tif 8.09e-08    0.000   0.000   0.014   0.986
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_18.tif 2.59e-06    0.056   0.000   0.001   0.943
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_19.tif 2.26e-06    0.000   0.000   0.022   0.978
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_20.tif 2.23e-06    0.000   0.000   0.008   0.992
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_21.tif 2.42e-06    0.000   0.000   0.013   0.987
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_22.tif 2.42e-09    0.000   0.010   0.015   0.975
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_23.tif 1.17e-10    0.000   1.000   0.000   0.000
    UNKNOWN BodyNR
./StraightWorm1/StraightenedWormtif_scan_24.tif 6.16e-09    0.000   0.002   0.007   0.990
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_25.tif 5.47e-06    0.000   0.000   0.004   0.996
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_26.tif 1.04e-05    0.000   0.000   0.004   0.996
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_27.tif 1.11e-08    0.000   0.002   0.005   0.993
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_28.tif 0.000127    0.000   0.000   0.000   1.000
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_29.tif 0.000192    0.000   0.000   0.017   0.982
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_30.tif 0.000173    0.000   0.000   0.048   0.952
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_31.tif 8.76e-06    0.000   0.010   0.021   0.969
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_32.tif 1.87e-10    0.000   0.949   0.016   0.035
    UNKNOWN BodyNR
./StraightWorm1/StraightenedWormtif_scan_33.tif 8.42e-08    0.000   0.003   0.004   0.993
    UNKNOWN TailNR
./StraightWorm1/StraightenedWormtif_scan_34.tif 0.000397    0.000   0.000   0.002   0.998
    UNKNOWN TailNR
                      BackgroundNR          BodyNR          HeadNR          TailNR
    BackgroundNR               0               0               0               0
          BodyNR               0               0               0               0
          HeadNR               0               0               0               0
          TailNR               0               0               0               0
         UNKNOWN               0               6               0              28

                      BackgroundNR          BodyNR          HeadNR          TailNR
    BackgroundNR         nan         nan         nan         nan
          BodyNR         nan         nan         nan         nan
          HeadNR         nan         nan         nan         nan
          TailNR         nan         nan         nan         nan
         UNKNOWN         0.05630         5.57337         0.76390         27.60642

Accuracy: nan 

Average accuracy (1 splits): nan

Original comment by dmarkeck...@gmail.com on 9 Mar 2011 at 4:12

GoogleCodeExporter commented 9 years ago 
The problem is not the -f switch.
wndchrm classify -f.005 /home/eckleyd/WormStageClassifier/AllStages/4stages.fit 
./15Cworms.fit 
worked well, no Seg Fault

Original comment by dmarkeck...@gmail.com on 9 Mar 2011 at 5:08

GoogleCodeExporter commented 9 years ago 
The confusion, similarity and distance matrixes need to be changed for datasets 
with unknown classes.
These matrixes report the status of the *test* set (not the training set), so 
the rows they contain should only reflect the classes defined in the test set.
If they are all unknown, then only the unknown row should be reported.
If the test set has only a subset of the classes in the training set defined, 
then only those should be reported.

Similarly, the accuracies should be based only on the samples in the *test* set 
with defined classes.

Original comment by i...@cathilya.org on 17 Mar 2011 at 5:50

GoogleCodeExporter commented 9 years ago 
The -f switch as well as the segfault were fixed with issue 25

Original comment by i...@cathilya.org on 22 Mar 2011 at 3:55