[Insight-users] Re: Journal submission gone wrong - probably
Luis Ibanez
luis.ibanez at kitware.com
Tue Jan 3 06:49:40 EST 2006
Hi Gaetan,
One easy way is to modify the Makefile for your
LaTeX build and include the image as a dependency
for the .dvi file.
Then you add the rule for building the image, as
simply running your executable with appropriate
parameters. It will have the additional advantage
that the readers of the paper will have access to
the exact parameters that you used for running
the executable.
Note that you should in any case, include the output
images in the source code submission. The testing
system will need them in order to perform the regression
testing. That is, the system at Kitware will run your
program again, it will generate output images and it
will compare them to the output images that you submit.
Ideally they should be exactly the same, but in practice
the ImageCompare program will have some tolerance on
the number of pixels that can be different between the
two images (your submitted output image and the output
image computed at the Insight Journal testing site).
Note in the CMakeLists.txt example on the template
submission, the use of the "ADD_TEST" and the use of
the ImageCompare command.
Please let us know if that addresses your question.
Thanks
Luis
-----------------------
Gaetan Lehmann wrote:
>
> Hi Luis,
>
> I think it's a very good idea to use the same images for the article
> and the tests. I have done that for the next revision.
> However, I don't know what to do with images generated by a program for
> the paper, like colored extrema on top of the input image. Should the
> Makefile run the program to generate the output image ? In that case,
> where will it found the program ? Should it be build by the Makefile ?
> Or should we have the output image with the other images in the
> archive, and add a test to verify the output image ?
>
> Regards,
>
> Gaetan
>
> On Fri, 23 Dec 2005 05:18:24 +0100, Luis Ibanez
> <luis.ibanez at kitware.com> wrote:
>
>>
>> Hi Gaetan,
>>
>>
>> 1) We checked the submission of Richard's paper,
>> it seems that there was a network connection problem
>> between the testing server at Kitware and the DSpace
>> server at UNC. Zack rerun the test and the dashboard
>> evaluation is now posted in the Insight Journal.
>>
>>
>> 2) About posting the source material of the paper, it is
>> fine if you find better to put it as a subdirectory
>> of the source code. I would prefer to have it independently
>> just because some papers may become large, specially if
>> they have images inserted, and because many users may
>> be interested in the source code more than in the source
>> text of the paper, so they would be penalized when downloading.
>>
>> On the other hand, one nice thing about your suggestion,
>> is that by putting the paper source text in the same
>> directory tree as the source code, the paper can easily
>> use the same images as the source code, and in that way
>> get closer to the ideal situation where the images in the
>> paper are *exactly* the same as the ones you are using
>> for running your code. This is probably the best way of
>> having *reproducible* papers.
>>
>>
>> So... to summarize, please feel free to submit the
>> source text + graphics of the paper as a subdirectory
>> of the source code submission.
>>
>> It is still preferable to have the PDF appart, so
>> readers can take a look at the paper before they
>> decide to invest time trying the source code.
>>
>>
>>
>> Thanks a lot for your support to the Insight Journal
>> and to ITK.
>>
>>
>>
>> Regards,
>>
>>
>> Luis
>>
>>
>>
>> -----------------------
>> Gaetan Lehmann wrote:
>>
>>> Hi Luis,
>>> Thanks for your useful report :-)
>>> The ImageCompare path will be fixed.
>>> The article is in the archive. It is called Article.pdf. But you're
>>> right, it should be outside the tar.gz. Because I've already done
>>> that for my previous submissions, I should have advise Richard to do
>>> that.
>>> I think it's nice to submit the sources of the paper, and I'm glad
>>> to see you also think so. However, I don't think that's useful to
>>> send the paper sources in a separated archive. It's just more work
>>> and more archives. Do you really find it useful ?
>>> About tests, I'm surprised to not see the results of the IJ
>>> dashboard. Is there a problem ?
>>> Finally, the second author (me) appear on the review page
>>> (http://insight-journal.org/view_reviews.php?back=index.php&pubid=62)
>>> but not on the first page.
>>> Gaetan
>>> On Thursday 22 December 2005 16:50, Luis Ibanez wrote:
>>>
>>>> Hi Richard,
>>>>
>>>>
>>>> Another detail,
>>>>
>>>> It seems that CMake is not searching for "ImageCompare" in the path.
>>>>
>>>> So you may want to add the following line to your CMakeLists.txt file
>>>>
>>>>
>>>>
>>>> FIND_PROGRAM(IMAGE_COMPARE ImageCompare)
>>>>
>>>>
>>>> and in the TEST lines replace "ImageCompare" with "${IMAGE_COMPARE}",
>>>> so the lines will look like:
>>>>
>>>>
>>>> ADD_TEST(VRMinFCompare ${IMAGE_COMPARE}
>>>> cthead1-vrminF.png
>>>> cthead1-vrminF-ref.png)
>>>>
>>>>
>>>>
>>>> Regards,
>>>>
>>>>
>>>> Luis
>>>>
>>>>
>>>> --------------------
>>>>
>>>> Richard Beare wrote:
>>>>
>>>>> Hi,
>>>>> I've just submitted something to the Insight Journal and suspect I've
>>>>> got it totally wrong. I've uploaded the archive created from Gaetan's
>>>>> template, which includes all the code, data and the article. I
>>>>> anticipated that it was going to be unpacked, and the article
>>>>> extracted, but it doesn't look as though this has happened, so I
>>>>> suspect I've made a mistake and should have handled the submission
>>>>> differently. I didn't see anywhere that would allow the upload of
>>>>> multiple files.
>>>>>
>>>>> If I have messed up severely then let me know what I should do to fix
>>>>> it. The handle is:
>>>>>
>>>>> http://hdl.handle.net/1926/153
>>>>
>>>>
>>>> _______________________________________________
>>>> Insight-users mailing list
>>>> Insight-users at itk.org
>>>> http://www.itk.org/mailman/listinfo/insight-users
>>>>
>>>>
>>>> ------------------------------------------------------------------------
>>>>
>>>>
>>>> _______________________________________________
>>>> Insight-users mailing list
>>>> Insight-users at itk.org
>>>> http://www.itk.org/mailman/listinfo/insight-users
>
>
>
>
More information about the Insight-users
mailing list