Mailman 3 April 2016 - scikit-image

Evaluating the predictive accuracy of the NB model
by potato_head April 16, 2016

April 16, 2016

Please guys, What am I doing wrong with using scikitlearn from nltk to check the accuracy of the naive bayes classifier? ...readFile definition not needed #divide the data into training and testing sets data = readFile('Data_test/') training_set = list_nltk[:2000000] testing_set = list_nltk[2000000:] #applied Bag of words as a way to select and extract feature count_vect = CountVectorizer() X_train_counts = count_vect.fit_transform(training_set.split('\n')) #apply tfd tf_transformer = TfidfTransformer(use_idf=False).fit(X_train_counts) X_train_tf = tf_transformer.transform(X_train_counts) #Train the data clf = MultinomialNB().fit(X_train_tf, training_set.split('\n')) #now test the accuracy of the naive bayes classifier test_data_features = count_vect.transform(testing_set) X_new_tfidf = tf_transformer.transform(test_data_features) predicted = clf.predict(X_new_tfidf)print "%.3f" % nltk.classify.accuracy(clf, predicted) The problem is when I print the nltk.classify.accuracy, it takes forever and I am suspecting this is because I have done something wrong but since I get no error, I can't figure out what it is that is wrong. I would really appreciate any pointer.

2 1

Re: profile_line
by Jaime Lopez Carvajal April 15, 2016

April 15, 2016

Hi Juan, Yeah, that is what I needed, no interpolation, so I will set it to 0. Thanks, Jaime On Friday, April 15, 2016 at 1:14:22 AM UTC-4, Juan Nunez-Iglesias wrote: > > Hey Jaime, > > Yes, if you set the interpolation order equal to 0, that resolves to > nearest neighbor interpolation, so it will pick the value from the nearest > pixel. The coordinates will still be fractional so that distances are > preserved, though. I hope that's still useful to you! =) > > Juan. > > On Fri, Apr 15, 2016 at 9:43 AM, Jaime Lopez Carvajal <jalo...(a)gmail.com > <javascript:>> wrote: > >> Hi Juan, >> >> Thanks, now I understand how it works. >> Do you know how can I get the gray values without interpolation, just the >> raw gray values taken from original image? >> >> Thanks in advance, Jaime >> >> On Thursday, April 14, 2016 at 6:41:43 PM UTC-4, Juan Nunez-Iglesias >> wrote: >>> >>> No, those aren't the correct coordinates. ;) >>> >>> profile_line ensures that the coordinates are spaced 1 pixel apart, and >>> then performs interpolation (the "order" argument to the function) to find >>> the values. >>> >>> On Fri, Apr 15, 2016 at 1:54 AM, Jaime Lopez Carvajal <jalo...(a)gmail.com >>> > wrote: >>> >>>> Sorry, the correct coordinates are: >>>> coords = np.array([[0, 0], >>>> >>>> [0, 0], >>>> [1, 1], >>>> [1, 2], >>>> [2, 3], >>>> [2, 3], >>>> [3, 4], >>>> [4, 5], >>>> [4, 6], >>>> [5, 6], >>>> [5, 7], >>>> [6, 8], >>>> [7, 9]]) >>>> >>>> >>>> On Thursday, April 14, 2016 at 11:42:01 AM UTC-4, Jaime Lopez Carvajal >>>> wrote: >>>>> >>>>> HI friends, >>>>> >>>>> I am having problems with *profile_line function*, because I want to >>>>> get gray values following some coordinates, but it is given to me wrong >>>>> gray values >>>>> that does not exist in the original image, like 128, 126 or 64. >>>>> >>>>> image = np.array([[170, 170, 70, 70, 70, 70, 70, 70, 70, 70], >>>>> [170, 170, 70, 70, 70, 70, 70, 70, 90, 70], >>>>> [170, 170, 170, 70, 70, 70, 70, 70, 70, 70], >>>>> [170, 170, 170, 70, 70, 70, 70, 70, 70, 70], >>>>> [170, 170, 170, 70, 70, 70, 70, 70, 70, 70], >>>>> [170, 170, 170, 170, 70, 70, 70, 70, 70, 70], >>>>> [110, 170, 170, 170, 70, 70, 70, 70, 70, 70], >>>>> [ 50, 170, 170, 170, 70, 70, 70, 70, 70, 10], >>>>> [ 50, 50, 170, 170, 70, 70, 70, 70, 10, 10], >>>>> [ 50, 110, 110, 110, 70, 70, 70, 70, 10, 10]], dtype=uint8) >>>>> >>>>> >>>>> coords = np.array([[70, 0], >>>>> >>>>> [70, 0], >>>>> [71, 1], >>>>> [71, 2], >>>>> [72, 3], >>>>> [72, 4], >>>>> [73, 4], >>>>> [74, 5], >>>>> [74, 6], >>>>> [75, 7], >>>>> [75, 8], >>>>> [76, 8], >>>>> [77, 9]]) >>>>> >>>>> >>>>> gray_values = profile_line(image,(0,0),(7,9)).astype(int) >>>>> >>>>> gray_values >>>>> >>>>> array([170, 170, 128, 126, 70, 70, 70, 70, 70, 70, 70, 64, 10]) >>>>> >>>>> >>>>> but the correct values would be: >>>>> >>>>> gray_values >>>>> >>>>> array([170, 170, 170, 70, 70, 70, 70, 70, 70, 70, 70, 70, 70]) >>>>> >>>>> >>>>> Do you know why this is happening? Any suggestion? >>>>> >>>>> >>>>> Thanks in advance, Jaime >>>>> >>>>> -- >>>> 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...(a)googlegroups.com. >>>> To post to this group, send email to scikit...(a)googlegroups.com. >>>> To view this discussion on the web, visit >>>> https://groups.google.com/d/msgid/scikit-image/28bd27cf-0761-4f73-89f9-61db… >>>> <https://groups.google.com/d/msgid/scikit-image/28bd27cf-0761-4f73-89f9-61db…> >>>> . >>>> >>>> For more options, visit https://groups.google.com/d/optout. >>>> >>> >>> -- >> 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...(a)googlegroups.com <javascript:>. >> To post to this group, send email to scikit...(a)googlegroups.com >> <javascript:>. >> To view this discussion on the web, visit >> https://groups.google.com/d/msgid/scikit-image/e9e7fcae-270b-44cf-8f48-e809… >> <https://groups.google.com/d/msgid/scikit-image/e9e7fcae-270b-44cf-8f48-e809…> >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > >

1 0

Re: profile_line
by Jaime Lopez Carvajal April 14, 2016

April 14, 2016

Hi Juan, Thanks, now I understand how it works. Do you know how can I get the gray values without interpolation, just the raw gray values taken from original image? Thanks in advance, Jaime On Thursday, April 14, 2016 at 6:41:43 PM UTC-4, Juan Nunez-Iglesias wrote: > > No, those aren't the correct coordinates. ;) > > profile_line ensures that the coordinates are spaced 1 pixel apart, and > then performs interpolation (the "order" argument to the function) to find > the values. > > On Fri, Apr 15, 2016 at 1:54 AM, Jaime Lopez Carvajal <jalo...(a)gmail.com > <javascript:>> wrote: > >> Sorry, the correct coordinates are: >> coords = np.array([[0, 0], >> >> [0, 0], >> [1, 1], >> [1, 2], >> [2, 3], >> [2, 3], >> [3, 4], >> [4, 5], >> [4, 6], >> [5, 6], >> [5, 7], >> [6, 8], >> [7, 9]]) >> >> >> On Thursday, April 14, 2016 at 11:42:01 AM UTC-4, Jaime Lopez Carvajal >> wrote: >>> >>> HI friends, >>> >>> I am having problems with *profile_line function*, because I want to >>> get gray values following some coordinates, but it is given to me wrong >>> gray values >>> that does not exist in the original image, like 128, 126 or 64. >>> >>> image = np.array([[170, 170, 70, 70, 70, 70, 70, 70, 70, 70], >>> [170, 170, 70, 70, 70, 70, 70, 70, 90, 70], >>> [170, 170, 170, 70, 70, 70, 70, 70, 70, 70], >>> [170, 170, 170, 70, 70, 70, 70, 70, 70, 70], >>> [170, 170, 170, 70, 70, 70, 70, 70, 70, 70], >>> [170, 170, 170, 170, 70, 70, 70, 70, 70, 70], >>> [110, 170, 170, 170, 70, 70, 70, 70, 70, 70], >>> [ 50, 170, 170, 170, 70, 70, 70, 70, 70, 10], >>> [ 50, 50, 170, 170, 70, 70, 70, 70, 10, 10], >>> [ 50, 110, 110, 110, 70, 70, 70, 70, 10, 10]], dtype=uint8) >>> >>> >>> coords = np.array([[70, 0], >>> >>> [70, 0], >>> [71, 1], >>> [71, 2], >>> [72, 3], >>> [72, 4], >>> [73, 4], >>> [74, 5], >>> [74, 6], >>> [75, 7], >>> [75, 8], >>> [76, 8], >>> [77, 9]]) >>> >>> >>> gray_values = profile_line(image,(0,0),(7,9)).astype(int) >>> >>> gray_values >>> >>> array([170, 170, 128, 126, 70, 70, 70, 70, 70, 70, 70, 64, 10]) >>> >>> >>> but the correct values would be: >>> >>> gray_values >>> >>> array([170, 170, 170, 70, 70, 70, 70, 70, 70, 70, 70, 70, 70]) >>> >>> >>> Do you know why this is happening? Any suggestion? >>> >>> >>> Thanks in advance, Jaime >>> >>> -- >> 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...(a)googlegroups.com <javascript:>. >> To post to this group, send email to scikit...(a)googlegroups.com >> <javascript:>. >> To view this discussion on the web, visit >> https://groups.google.com/d/msgid/scikit-image/28bd27cf-0761-4f73-89f9-61db… >> <https://groups.google.com/d/msgid/scikit-image/28bd27cf-0761-4f73-89f9-61db…> >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > >

2 1

profile_line
by Jaime Lopez Carvajal April 14, 2016

April 14, 2016

HI friends, I am having problems with *profile_line function*, because I want to get gray values following some coordinates, but it is given to me wrong gray values that does not exist in the original image, like 128, 126 or 64. image = np.array([[170, 170, 70, 70, 70, 70, 70, 70, 70, 70], [170, 170, 70, 70, 70, 70, 70, 70, 90, 70], [170, 170, 170, 70, 70, 70, 70, 70, 70, 70], [170, 170, 170, 70, 70, 70, 70, 70, 70, 70], [170, 170, 170, 70, 70, 70, 70, 70, 70, 70], [170, 170, 170, 170, 70, 70, 70, 70, 70, 70], [110, 170, 170, 170, 70, 70, 70, 70, 70, 70], [ 50, 170, 170, 170, 70, 70, 70, 70, 70, 10], [ 50, 50, 170, 170, 70, 70, 70, 70, 10, 10], [ 50, 110, 110, 110, 70, 70, 70, 70, 10, 10]], dtype=uint8) coords = np.array([[70, 0], [70, 0], [71, 1], [71, 2], [72, 3], [72, 4], [73, 4], [74, 5], [74, 6], [75, 7], [75, 8], [76, 8], [77, 9]]) gray_values = profile_line(image,(0,0),(7,9)).astype(int) gray_values array([170, 170, 128, 126, 70, 70, 70, 70, 70, 70, 70, 64, 10]) but the correct values would be: gray_values array([170, 170, 170, 70, 70, 70, 70, 70, 70, 70, 70, 70, 70]) Do you know why this is happening? Any suggestion? Thanks in advance, Jaime

2 2

Fwd: [Scikit-learn-general] Binary wheel packages for Linux are coming
by Stéfan van der Walt April 14, 2016

April 14, 2016

Hi everyone Please see Olivier's email below. The same holds for scikit-image, so pip install scikit-image should now do a binary install on (many) Linux systems. Please thank Nathaniel Smith (who wrote the manylinux PEP and worked on pypi/pip integration), Robbert McGibbon (who wrote the tools to include libraries in the wheels), Matthew Brett (who set up build infrastructure & tested), and Olivier (who patched pypi/pip and tested). Best regards Stéfan ---------- Forwarded message ---------- From: Olivier Grisel Date: 13 April 2016 at 14:52 Subject: Re: [Scikit-learn-general] Binary wheel packages for Linux are coming OpenBLAS 0.2.18 has been released yesterday with many fixes several bugs found via testing the scipy stack on: https://build.openblas.net/ Matthew Brett did a full-rebuild of BLAS dependent projects (including numpy, scipy and scikit-learn) and the resulting wheels are being uploaded to PyPI. On most Linux distros you can now already do: python -m pip install --upgrade pip # to upgrade to 8.1 or later python -m pip install numpy scipy scikit-learn And get the stack installed with an optimized BLAS/LAPACK without having to compile anything. Next I will work on updating our CI configuration to build and test those wheels on our master branch and for future release tags. Best, -- Olivier ------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z _______________________________________________ Scikit-learn-general mailing list Scikit-learn-general(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

1 0

Please share your use cases
by Stéfan van der Walt April 13, 2016

April 13, 2016

Hi, everyone I am giving a talk on scikit-image soon, and would like to highlight how it is being used in practice. Please share your own applications with links to images or plots, or let me know if you spot someone else doing something cool! Thanks, Stéfan

1 0

suppress UserWarning
by Gustavo Goretkin April 11, 2016

April 11, 2016

I'm on skimage.__version__ '0.12.3' and I see warnings like `UserWarning: test.png is a low contrast image` and it gets printed every time, not just once. How can I suppress this warning? I am not even sure if skimage is responsible for printing this. Thanks, Gustavo

1 1

[JOB] Fwd: [Comp-neuro] Image Analysis Job Opportunity at Salk Institute
by Ariel Rokem April 9, 2016

April 9, 2016

Forgive cross-posting. This might be of interest to denizens of this list. ---------- Forwarded message ---------- From: Uri Manor <umanor(a)salk.edu> Date: Fri, Apr 8, 2016 at 3:56 PM Subject: [Comp-neuro] Image Analysis Job Opportunity at Salk Institute To: "comp-neuro(a)neuroinf.org" <comp-neuro(a)neuroinf.org> Hi everyone, Please spread the word far and wide! Thanks! WAITT ADVANCED BIOPHOTONICS CORE Job Opportunity – Image Analysis Scientist The Waitt Advanced Biophotonics Center has been established at the Salk Institute for Biological Studies as a state-of-the-art research hub to provide advanced and emerging imaging modalities to enable unprecedented insights into the inner workings of cells, tissues and organisms. We are seeking a highly motivated and experienced candidate in the field of computational image analysis to serve as an Image Analysis Scientist for the Biophotonics Core Facility. The Image Analysis Scientist will have a principal responsibility to collaborate with the other light and electron microscopy researchers in the Biophotonics Core and the Salk Institute to provide specific expertise for developing custom computer algorithms for complex and novel analyses of raw imaging data, computer data management and developing algorithms for microscopy operations. The Image Analysis Scientist will also develop and publish novel imaging and image analysis techniques in collaboration with the Core Director and Salk faculty. The Image Analysis Scientist will apply existing image processing software to 2D and 3D images and videos, and will also integrate data across different platforms (e.g. correlative EM). Finally, under the supervision and training of the Core Director, the Image Analysis Scientist will become familiar with the microscopes used in the facility, and develop expertise in using the equipment to acquire imaging data optimized for image analysis. Along with the Core Director, the Image Analysis Scientist will train Salk scientists on how to properly acquire high quality quantitative microscopy image data. This is a multi-faceted job that involves the following components: • Image Processing & Data Analysis - With the support of the Core Director, the incumbent will participate in the collaborative development of custom image data analysis and algorithm solutions. • Research Design & Participation - aiding Director and independently assisting users in design of imaging experiments. Educate facility users in the use of facility instrumentation and methods used therein for optimal acquisition of imaging data. • Data Collection & Presentation - collecting and managing data and helping to prepare data for publication & presentation. • User Training - guide and assist Salk researchers in the use of existing image acquisition and processing software. To Apply: Send CV to umanor(a)salk.edu or http://www.salk.edu/about/careers/ (Job # F142) Uri Manor, Ph.D. DIRECTOR WAITT ADVANCED BIOPHOTONICS CORE PH (858) 453-4100 x2119 *M *(314) 406-7618 E umanor(a)salk.edu <tberggren(a)salk.edu> W http://bpho.salk.edu <tberggren(a)salk.edu> Salk Institute for Biological Studies 10010 N Torrey Pines Rd • La Jolla, CA 92037 WWW.SALK.EDU <http://www.salk.edu> _______________________________________________ Comp-neuro mailing list Comp-neuro(a)neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro

1 0

Re: New core team member
by Egor Panfilov April 5, 2016

April 5, 2016

Thank you very much, lady and gentlemen! ;) It is a great pleasure for me to work with you side by side on such a strong project! Kind regards, Egor Panfilov

1 0

Re: New core team member
by Emmanuelle Gouillart April 5, 2016

April 5, 2016

Welcome Egor! It's awesome to see the core team grow thanks to dedicated and enthusiastic new members like you! Emma On Tue, Apr 05, 2016 at 06:04:30PM +1000, Juan Nunez-Iglesias wrote: > Welcome, Egor! And I'll reiterate StÃ©fan's statement: thank you very much for > all your hard work on github so far! It's been impressive. =D > On Tue, Apr 5, 2016 at 12:47 PM, StÃ©fan van der Walt <stefanv(a)berkeley.edu> > wrote: > Dear scikit-image community > It is my pleasure to welcome Egor Panfilov (@soupault) to the core > team!Â As you may have seen on GitHub, Egor has been hard at working > triaging, reviewing, and contributing. > Thank you for your hard work so far, Egor, and we look forward to your > future contributions! > Best regards > StÃ©fan

1 0