Mailman 3 November 2013 - scikit-image

Re: SLIC Module in SciKit
by Jeremy Dunne 18 Nov '13

18 Nov '13

Stefan, Well that seems to have solved it...thank you very much....atleast I have a new error ..... resizing rgb and nir images RGB image name \Users\jeremydunn\Desktop\GroundSure\Algorithm\Wales_Drew\Tadas\Area2\Imagery\gm0F6F0.jpg Traceback (most recent call last): File "C:\Users\jeremydunn\Desktop\GroundSure\Algorithm\Wales_Drew\Tadas\Algorithm\run3.py", line 79, in <module> training_1 = TrainImage(trainingrgb, trainingnir, object_size, size_to_filter, jappos = trainingpos, japneg = trainingneg) File "C:\Users\jeremydunn\Desktop\GroundSure\Algorithm\Wales_Drew\Tadas\Algorithm\algorithm.py", line 51, in __init__ self.pixres,self.gt,self.prj) = og.initialize(filename,nirfile,jappos,japneg) File "C:\Users\jeremydunn\Desktop\GroundSure\Algorithm\Wales_Drew\Tadas\Algorithm\ogr_gdal.py", line 171, in initialize proc = subprocess.Popen(['gdal_translate','-of','JPEG','-projwin',str(extents[0]),str(extents[1]),str(extents[2]),str(extents[3]),filename,'tmp_rgb.jpg']) File "C:\Python27\ArcGIS10.2\lib\subprocess.py", line 679, in __init__ errread, errwrite) File "C:\Python27\ArcGIS10.2\lib\subprocess.py", line 896, in _execute_child startupinfo) WindowsError: [Error 2] The system cannot find the file specified Thank you for the help, Cheers. Jeremy. On Monday, 18 November 2013 16:28:40 UTC, Stefan van der Walt wrote: > > On Mon, Nov 18, 2013 at 6:15 PM, Stéfan van der Walt <ste...(a)sun.ac.za<javascript:>> > wrote: > > from skimage.morphology import slic > > Sorry, where is my mind: > > from skimage.segmentation import slic > > Stéfan >

2 1

Re: SLIC Module in SciKit
by Stéfan van der Walt 18 Nov '13

18 Nov '13

On Mon, Nov 18, 2013 at 6:15 PM, Stéfan van der Walt <stefan(a)sun.ac.za> wrote: > from skimage.morphology import slic Sorry, where is my mind: from skimage.segmentation import slic Stéfan

1 0

Re: SLIC Module in SciKit
by Jeremy Dunne 18 Nov '13

18 Nov '13

Hi Stefan, Ok I have tried this a couple of ways but still the same error..... Error 1 Traceback (most recent call last): File "C:\Users\jeremydunn\Desktop\GroundSure\Algorithm\Wales_Drew\Tadas\Algorithm\run3.py", line 2, in <module> from algorithm import * File "C:\Users\jeremydunn\Desktop\GroundSure\Algorithm\Wales_Drew\Tadas\Algorithm\algorithm.py", line 7, in <module> from skimage.segmentation.slic import slic ImportError: No module named slic Error 2 Traceback (most recent call last): File "C:\Users\jeremydunn\Desktop\GroundSure\Algorithm\Wales_Drew\Tadas\Algorithm\run3.py", line 2, in <module> from algorithm import * File "C:\Users\jeremydunn\Desktop\GroundSure\Algorithm\Wales_Drew\Tadas\Algorithm\algorithm.py", line 7, in <module> from skimage.morphology import slic ImportError: cannot import name slic Am I missing something basic here ? Probably. Cheers. Jeremy. On Monday, 18 November 2013 16:15:28 UTC, Stefan van der Walt wrote: > > On Mon, Nov 18, 2013 at 6:09 PM, Jeremy Dunne <jdun...(a)gmail.com<javascript:>> > wrote: > > This is everything that I am importing, does order matter ? > > No, but there is no such stand-alone module as "slic". You have to > import it as follows: > > from skimage.morphology import slic > > Stéfan >

1 0

Re: SLIC Module in SciKit
by Jeremy Dunne 18 Nov '13

18 Nov '13

Hi Stefan, This is everything that I am importing, does order matter ? #!/usr/bin/python import os import numpy as np from scipy import ndimage as ndi from skimage.segmentation.boundaries import find_boundaries import scipy.ndimage.measurements as msr import slic as slic import skimage.color as skic from sklearn import svm from sklearn.preprocessing import StandardScaler from sklearn.cross_validation import StratifiedKFold from sklearn.grid_search import GridSearchCV import scipy.ndimage.morphology as morph from skimage.filter import threshold_otsu from sklearn import decomposition from scipy.ndimage import convolve import gabor as gb import ogr_gdal as og import matplotlib.pyplot as plt from params import * Alot but the only issue is slic, if I comment it out code run until it needs slic. Is it supported in skimage ? Error Traceback (most recent call last): File "C:\Users\jeremydunn\Desktop\GroundSure\Algorithm\Wales_Drew\Tadas\Algorithm\run3.py", line 2, in <module> from algorithm import * File "C:\Users\jeremydunn\Desktop\GroundSure\Algorithm\Wales_Drew\Tadas\Algorithm\algorithm.py", line 7, in <module> import slic as slic ImportError: No module named slic Cheers. Jeremy. On Monday, 18 November 2013 15:51:34 UTC, Stefan van der Walt wrote: > > On Mon, Nov 18, 2013 at 5:34 PM, Jeremy Dunne <jdun...(a)gmail.com<javascript:>> > wrote: > > Nope that's not the issue. Although I checked to make sure. From the > code it > > is called as below > > > > import slic as slic > > You cannot import slic by itself, you need to import > ``skimage.segmentation.slic``, or just import segmentation and use > ``segmentation.slic``. > > Stéfan >

2 1

Re: SLIC Module in SciKit
by Jeremy Dunne 18 Nov '13

18 Nov '13

Hi Juan, Nope that's not the issue. Although I checked to make sure. From the code it is called as below import slic as slic This is the only module it bulks at, all the others are fine. There are a lot but they all seem to be fine. #!/usr/bin/python import os import numpy as np from scipy import ndimage as ndi from skimage.segmentation.boundaries import find_boundaries import scipy.ndimage.measurements as msr import slic as slic import skimage.color as skic from sklearn import svm from sklearn.preprocessing import StandardScaler from sklearn.cross_validation import StratifiedKFold from sklearn.grid_search import GridSearchCV import scipy.ndimage.morphology as morph from skimage.filter import threshold_otsu from sklearn import decomposition from scipy.ndimage import convolve import gabor as gb import ogr_gdal as og import matplotlib.pyplot as plt from params import * Cheers. Jeremy. On Monday, 18 November 2013 13:08:33 UTC, Juan Nunez-Iglesias wrote: > > Hi Jeremy, > > is this just an issue of case? ie "slic" vs "SLIC"? The correct import > line should be: > > from skimage.segmentation import slic > > I hope this helps! > > Juan. > > On Monday, November 18, 2013, Jeremy Dunne wrote: > >> Morning All, >> >> >> >> I have been trying to port my linux python code to a windows x32 >> environment and am very close however I just can’t seem to find a way of >> Python 2.7 recognising the SLIC Module that is within sci-image. Having >> used Chris Gohlke’s Windows Extensions everything works up until my code >> looks to import SLIC from SciKit. >> >> >> >> Error : No module SLIC >> >> >> >> Anyone else had this issue or solved this problem ? >> >> >> >> Cheers >> >> >> >> Jeremy. >> >> -- >> You received this message because you are subscribed to the Google Groups >> "scikit-image" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to scikit-image+unsubscribe(a)googlegroups.com. >> For more options, visit https://groups.google.com/groups/opt_out. >> >

2 1

SLIC Module in SciKit
by Jeremy Dunne 18 Nov '13

18 Nov '13

Morning All, I have been trying to port my linux python code to a windows x32 environment and am very close however I just can’t seem to find a way of Python 2.7 recognising the SLIC Module that is within sci-image. Having used Chris Gohlke’s Windows Extensions everything works up until my code looks to import SLIC from SciKit. Error : No module SLIC Anyone else had this issue or solved this problem ? Cheers Jeremy.

2 1

Re: scikit-image paper
by Johannes Schönberger 18 Nov '13

18 Nov '13

> Currently my two top candidates then are JMLR MLOSS and PLOS ONE. +1 > Let's get started on some content! +1, but I must mention that I might be short of time over the coming weeks as I am about to move to another continent, so unfortunately I have a ton of other things to do :-) > I think we would do well to emphasize the following in the paper: > > - Pythonic API > - Building block for reproducible research, use in academic world (we > need a list of publications of work that used skimage) > - Educational aspects (skimage.novice) > - Google Summer of Code contributions > - Use in industry (do we have examples here?) > - ... other ideas? - Extensive documentation (although the docs need some more „love“ for 0.10) - Easy to use API for prototyping and testing algorithms (probably covered by "Pythonic API“) I have mainly used skimage for prototyping, but did not directly mention it in a publication. Johannes

1 0

Re: scikit-image paper
by Stéfan van der Walt 18 Nov '13

18 Nov '13

On Mon, Nov 18, 2013 at 10:47 AM, Juan Nunez-Iglesias <jni.soma(a)gmail.com> wrote: > If we are going to worry about JIF, PLOS ONE has 3.7, and does accept papers > from all disciplines. (You will also find a Scholarly Kitchen post about how > a JIF of 3.7 is a sign that PLOS and open access are doomed.) Thanks, I won't take Scholarly Kitchen too seriously then. I know that impact factors are a lousy way (by itself) to determine where to publish, but that's a whole other kettle of fish. > Currently my two top candidates then are JMLR MLOSS and PLOS ONE. (The > latter does charge $1300 US per article.) Open access fees should not pose too much of a problem: http://library.sun.ac.za/English/services/oa/Pages/su-oafund.aspx Let's get started on some content! I think we would do well to emphasize the following in the paper: - Pythonic API - Building block for reproducible research, use in academic world (we need a list of publications of work that used skimage) - Educational aspects (skimage.novice) - Google Summer of Code contributions - Use in industry (do we have examples here?) - ... other ideas? Stéfan

1 0

Re: scikit-image paper
by Juan Nunez-Iglesias 18 Nov '13

18 Nov '13

Hey Stéfan, On Mon, Nov 18, 2013 at 6:13 PM, Stéfan van der Walt <stefan(a)sun.ac.za>wrote: > > - F1000 Research > > I'm a bit concerned by what I read here: > > http://scholarlykitchen.sspnet.org/2013/01/15/pubmed-and-f1000-research-unc… Scholarly Kitchen have it in for open access in general and seem to think there is nothing wrong with the business-as-usual model of scientific publishing. I imagine that this is not true of the present list of authors, but am happy to be corrected. Just have a quick browse of their past posts and try to find something positive about OA. You can just read a few of Fernando Perez's blog posts, particularly the recent one about the Sloan/Moore data science initiative, to find out what I think of business-as-usual science and publishing. Quick rebuttals of the post's two main points: "unclear standards applied unevenly" "cynical and confusing mélange of incomplete editorial practices" Yeah, welcome to peer review. This is true of all journals — the fact that it's not open doesn't mean it's not there. "Yet, these rejected articles continue to be published on the F1000 Research site." It's called a preprint. arXiv papers don't get taken down when they don't pass peer review (ditto for closed github PRs). Why should these? They are clearly marked as Rejected. F1000 Research is an experiment in openness in publishing and I think it's a worthwhile one. Having said these things, it's definitely a risk publishing there. (My impression though is that scikit-image as a library speaks for itself and the choice of journal is a choice of what model we want to support, rather than which journal will be more prestigious.) *But, *more importantly, it seems that F1000 Research is indeed slanted towards the biological sciences; I was under the impression that it was a catchall journal. So, it might not be the best fit. [rant over] =) > - JMLR MLOSS [sklearn published here] > > Impact factor is about 3.4, which at least gives some indication that > its content is valued. > If we are going to worry about JIF, PLOS ONE has 3.7, and *does* accept papers from all disciplines. (You will also find a Scholarly Kitchen post about how a JIF of 3.7 is a sign that PLOS and open access are doomed.) Currently my two top candidates then are JMLR MLOSS and PLOS ONE. (The latter does charge $1300 US per article.) Let's stick to using wrapped lines, otherwise editing using our normal > configurations becomes painful. > Agreed. Juan.

1 0

Re: scikit-image paper
by Stéfan van der Walt 18 Nov '13

18 Nov '13

On Mon, Nov 18, 2013 at 3:23 AM, Juan Nunez-Iglesias <jni.soma(a)gmail.com> wrote: > * This and Johannes's email sparked a discussion about possible homes for > it. Current suggestions: > - F1000 Research I'm a bit concerned by what I read here: http://scholarlykitchen.sspnet.org/2013/01/15/pubmed-and-f1000-research-unc… > - JMLR MLOSS [sklearn published here] Impact factor is about 3.4, which at least gives some indication that its content is valued. > - Journal on Image and Video Processing (jivp.eurasipjournals.com) And about 0.5 for this one. Let's stick to using wrapped lines, otherwise editing using our normal configurations becomes painful. Stéfan

1 0