From benalayaines at yahoo.fr  Wed May  4 18:21:32 2016
From: benalayaines at yahoo.fr (Ben Alaya Ines)
Date: Wed, 4 May 2016 22:21:32 +0000 (UTC)
Subject: [Neuroimaging] (no subject)
References: <291626549.1253649.1462400492979.JavaMail.yahoo.ref@mail.yahoo.com>
Message-ID: <291626549.1253649.1462400492979.JavaMail.yahoo@mail.yahoo.com>

I look for software that allow me to reconstruct the FOD and i don't have diffusion-weighted images. I will use the script?matlab of challenge ISBI'12 to generate the diffusion signal.
Best regards.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160504/3611edb9/attachment.html>

From arokem at gmail.com  Thu May  5 11:23:00 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Thu, 5 May 2016 08:23:00 -0700
Subject: [Neuroimaging] Journal articles based on PRs
In-Reply-To: <CA+X4w0x63ZomeUZcrTRwhQY6jrveE9QKT6CQcKjDYj4v_V7dyg@mail.gmail.com>
References: <CA+X4w0z1n===KzQTeBVjAog6PrfFk+4o4jVgdvR18T8go8_4Mw@mail.gmail.com>
 <CABDkGQkbLhi+Lp4vwfmbAF0Q-pX93gBvvUdzU6w=bonBQzirjw@mail.gmail.com>
 <CA+X4w0x63ZomeUZcrTRwhQY6jrveE9QKT6CQcKjDYj4v_V7dyg@mail.gmail.com>
Message-ID: <CA+X4w0zq-O7dpiB1+PMrq_SsmMbSh-ssFFzhS1_1M0Fq_dwPrQ@mail.gmail.com>

To answer my own question (in ouroboros fashion):

On Mon, Apr 25, 2016 at 4:49 PM, Ariel Rokem <arokem at gmail.com> wrote:

>
>
> On Mon, Apr 25, 2016 at 4:33 PM, St?fan van der Walt <stefanv at berkeley.edu
> > wrote:
>
>> On Tue, 12 Apr 2016 at 08:27 Ariel Rokem <arokem at gmail.com> wrote:
>>
>>> In a conversation I had with Rafael recently, he mentioned to me the
>>> Journal of Open Research Software (
>>> http://openresearchsoftware.metajnl.com/) that publishes articles about
>>> open-source research software, and proposed this as a good place to publish
>>> software contributions in our community.
>>>
>>
>> Karthik Ram recently told me about
>>
>> http://joss.theoj.org/about
>>
>> I would like to hear what others think of this journal (tl;dr: it takes
>> about 1-3 hrs to prepare a paper for publication in their peer-reviewed
>> journal).
>>
>
> Interesting. Looks like it's not quite up and running yet. Do you know if
> they are planning a mechanism whereby PRs could be counted as distinct
> contributions?
>

Seems that this is now up and running:
http://www.arfon.org/announcing-the-journal-of-open-source-software

And: https://github.com/openjournals/joss/issues/52


>
>
>> St?fan
>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160505/079d425e/attachment.html>

From arokem at gmail.com  Thu May  5 11:26:28 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Thu, 5 May 2016 08:26:28 -0700
Subject: [Neuroimaging] Journal articles based on PRs
In-Reply-To: <20160426052739.GP3689071@phare.normalesup.org>
References: <CA+X4w0z1n===KzQTeBVjAog6PrfFk+4o4jVgdvR18T8go8_4Mw@mail.gmail.com>
 <CABDkGQkbLhi+Lp4vwfmbAF0Q-pX93gBvvUdzU6w=bonBQzirjw@mail.gmail.com>
 <20160426052739.GP3689071@phare.normalesup.org>
Message-ID: <CA+X4w0wo0-FNk_TgYSRCMH2J-cnrz94NzgMJcxcSag1hju9sqg@mail.gmail.com>

And to answer Gael's concern:

On Mon, Apr 25, 2016 at 10:27 PM, Gael Varoquaux <
gael.varoquaux at normalesup.org> wrote:

> On Mon, Apr 25, 2016 at 11:33:15PM +0000, St?fan van der Walt wrote:
> > I would like to hear what others think of this journal (tl;dr: it takes
> about
> > 1-3 hrs to prepare a paper for publication in their peer-reviewed
> journal).
>
> 1-3 hrs to prepare a paper! Is that a good thing? It takes me more than 3
> hours to peer review a publications. It takes me at least an hour to read
> one. Don't we already have too many publications of low quality?


I think this assumes that you have already written such high-quality
documentation of the software that an additional thoughtful high-quality
manuscript is redundant information, and wouldn't be particularly useful
anyway. That is, if your software is not up to snuff, you're going to have
much more than 1-3 hours of work to prepare *your software* for publication
in this journal.



> G
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160505/e7b0640f/attachment.html>

From bertrand.thirion at inria.fr  Fri May  6 17:20:47 2016
From: bertrand.thirion at inria.fr (bthirion)
Date: Fri, 6 May 2016 23:20:47 +0200
Subject: [Neuroimaging] Nilearn CS in Paris, June 8-10 in Paris
Message-ID: <572D0AAF.8020209@inria.fr>

     Dear Nipyers,

     We're planning new coding sprint in Paris in about one month. You 
can find information here: 
https://github.com/nilearn/nilearn/wiki/June-2016-Sprint
     Please do not hesitate to propose some additional ideas and let me 
know if you wish to join us.
     Best,

Bertrand

From etienne.roesch at gmail.com  Tue May 10 05:30:58 2016
From: etienne.roesch at gmail.com (Etienne B. Roesch)
Date: Tue, 10 May 2016 09:30:58 +0000
Subject: [Neuroimaging] 2nd announcement,
 Summer School "Advanced Scientific Programming in Python" in
 Reading, UK, September 5--11, 2016
Message-ID: <CAPX-MzCh96UGWOjqP7CdLDSzU+bgB1aZgOoJFZr8aT8mL7qKjg@mail.gmail.com>

*Advanced Scientific Programming in Python*

a Summer School by the G-Node, and the Centre for Integrative Neuroscience
and Neurodynamics, School of Psychology and Clinical Language Sciences,
University of Reading, UK

Scientists spend more and more time writing, maintaining, and debugging
software. While techniques for doing this efficiently have evolved, only
few scientists have been trained to use them. As a result, instead of doing
their research, they spend far too much time writing deficient code and
reinventing the wheel. In this course we will present a selection of
advanced programming techniques and best practices which are standard in
the industry, but especially tailored to the needs of a programming
scientist. Lectures are devised to be interactive and to give the students
enough time to acquire direct hands-on experience with the materials.
Students will work in pairs throughout the school and will team up to
practice the newly learned skills in a real programming project ? an
entertaining computer game.

We use the Python programming language for the entire course. Python works
as a simple programming language for beginners, but more importantly, it
also works great in scientific simulations and data analysis. We show how
clean language design, ease of extensibility, and the great wealth of open
source libraries for scientific computing and data visualization are
driving Python to become a standard tool for the programming scientist.

This school is targeted at Master or PhD students and Post-docs from all
areas of science. Competence in Python or in another language such as Java,
C/C++, MATLAB, or Mathematica is absolutely required. Basic knowledge of
Python and of a version control system such as git, subversion, mercurial,
or bazaar is assumed. Participants without any prior experience with Python
and/or git should work through the proposed introductory material before
the course.

We are striving hard to get a pool of students which is international and
gender-balanced.

You can apply online: https://python.g-node.org



*Application deadline: 23:59 UTC, May 15, 2016. Be sure to read the FAQ
before applying. *

Participation is for free, i.e. no fee is charged! Participants however
should take care of travel, living, and accommodation expenses by
themselves. Travel grants may be available.


*Date & Location *
September 5?11, 2016. Reading, UK

*Program *
- Best Programming Practices
? Best practices for scientific programming
? Version control with git and how to contribute to open source projects
with GitHub
? Best practices in data visualization

- Software Carpentry
? Test-driven development
? Debugging with a debugger
? Profiling code

- Scientific Tools for Python
? Advanced NumPy

- Advanced Python
? Decorators
? Context managers
? Generators

- The Quest for Speed
? Writing parallel applications
? Interfacing to C with Cython
? Memory-bound problems and memory profiling
? Data containers: storage and fast access to large data

- Practical Software Development
? Group project

*Preliminary Faculty *
? Francesc Alted, freelance consultant, author of PyTables, Spain
? Pietro Berkes, Enthought Inc., Cambridge, UK
? Zbigniew J?drzejewski-Szmek, Krasnow Institute, George Mason University,
Fairfax, VA, USA
? Eilif Muller, Blue Brain Project, ?cole Polytechnique F?d?rale de
Lausanne, Switzerland
? Rike-Benjamin Schuppner, Institute for Theoretical Biology,
Humboldt-Universit?t zu Berlin, Germany
? Bartosz Tele?czuk, European Institute for Theoretical Neuroscience, CNRS,
Paris, France
? St?fan van der Walt, Berkeley Institute for Data Science, UC Berkeley,
CA, USA
? Nelle Varoquaux, Centre for Computational Biology Mines ParisTech,
Institut Curie, U900 INSERM, Paris, France
? Tiziano Zito, freelance consultant, Germany

*Organizers *
For the German Neuroinformatics Node of the INCF (G-Node) Germany:
? Tiziano Zito, freelance consultant, Germany
? Zbigniew J?drzejewski-Szmek, Krasnow Institute, George Mason University,
Fairfax, USA
? Jakob Jordan, Institute of Neuroscience and Medicine (INM-6),
Forschungszentrum J?lich GmbH, Germany

For the Centre for Integrative Neuroscience and Neurodynamics, School of
Psychology and Clinical Language Sciences, University of Reading UK:
? Etienne Roesch, Centre for Integrative Neuroscience and Neurodynamics,
University of Reading, UK

*Website*: https://python.g-node.org
*Contact*: python-info at g-node.org


Kind regards,

Etienne

-----
Dr. Etienne B. Roesch
Lecturer in Cognitive Science
University of Reading
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160510/9c521dba/attachment.html>

From alexandre.gramfort at telecom-paristech.fr  Wed May 11 03:59:36 2016
From: alexandre.gramfort at telecom-paristech.fr (Alexandre Gramfort)
Date: Wed, 11 May 2016 09:59:36 +0200
Subject: [Neuroimaging] [ANN] MNE-Python 0.12
Message-ID: <CADeotZoTuVV+FEmmxbA1cqo7wtyoSTPm=CQvfRZwxVTFN4r=qA@mail.gmail.com>

Hi,

We are pleased to announce the new 0.12 release of MNE-Python. This release
comes with many improvements to usability, visualization and documentation
and bug fixes.

A few highlights:

   -

   We entirely revamped our documentation at the MNE website with a new
   easy-to-follow structure, have a look at http://martinos.org/mne and let
   us know if you would like to read more on a particular topic. See eg.
   http://martinos.org/mne/stable/tutorials.html
   -

   We introduced annotations for marking arbitrary segments of raw data.
   This can be used in order to annotate M/EEG recordings with naturalistic
   stimuli or for rejecting bad segments of data. See
   http://martinos.org/mne/dev/auto_tutorials/plot_brainstorm_auditory.html
   for an example.
   -

   Added the ability to create animations/movies of sensor topographies.
   See Evoked.animate_topomap method.
   -

   We now have movement compensation for Maxwell-Filter
   -

   We have new short-hand plotting function for showing sensor positions
   and layouts.
   -

   We now explicitly support ECoG data with a specific ecog channel type.
   -

   Evoked activity (as a butterfly time series) and corresponding topomaps
   can now be shown in one plot with `Evoked.plot_joint()` for spatio-temporal
   brain dynamics
   -

   Support for reading and estimation of fixed-position dipole time courses
   (similar to Elekta ``xfit``)
   -

   New mne.io.read_raw_cnt function for reading Neuroscan CNT files


Notable API changes:

   -

   To unify in-place modification vs. copying API, the `copy` parameter was
   deprecated for all MNE object methods and will be removed in a later
   version; instead, `inst.copy().method()` is to be used. Also, all object
   methods now return `self`, allowing reliable chaining (e.g. `raw_resampled
   = raw.copy().filter(1).resample(100)`)
   -

   Generalization Across Time now supports custom predict functions, e.g.
   predicting probabilities rather than classes, via the `predict_method`
   keyword argument; and an option was added to score either across or within
   folds via the `predict_mode` keyword argument.
   -

   We now have additional decimation parameters for time-frequency methods
   -

   When estimating covariance from raw data, the same regularization
   methods can be used as for estimating the covariance from epoched data.
   -

   From now on ECG, EOG and EMG channels are shown by default  in butterfly
   plots


For a full list of improvements and API changes, see:

http://martinos.org/mne/stable/whats_new.html#version-0-12

To install the latest release the following command should do the job:

pip install --upgrade --user mne

As usual we welcome your bug reports, feature requests, critiques and

contributions.

Some links:

- https://github.com/mne-tools/mne-python (code + readme on how to install)

- http://martinos.org/mne/stable/ (full MNE documentation)

Follow us on Twitter: https://twitter.com/mne_python

Regards,

The MNE-Python developers

People who contributed to this release with their number of commits:

The committer list for this release is the following (preceded by

number of commits):


   -

   348 Eric Larson
   -

   347 Jaakko Leppakangas
   -

   157 Alexandre Gramfort
   -

   139 Jona Sassenhagen
   -

   67 Jean-Remi King
   -

   32 Chris Holdgraf
   -

   31 Denis A. Engemann
   -

   30 Mainak Jas
   -

   16 Christopher J. Bailey
   -

   13 Marijn van Vliet
   -

   10 Mark Wronkiewicz
   -

   9 Teon Brooks
   -

   9 kaichogami
   -

   8 Cl?ment Moutard
   -

   5 Camilo Lamus
   -

   5 mmagnuski
   -

   4 Christian Brodbeck
   -

   4 Daniel McCloy
   -

   4 Yousra Bekhti
   -

   3 Fede Raimondo
   -

   1 Jussi Nurminen
   -

   1 MartinBaBer
   -

   1 Mikolaj Magnuski
   -

   1 Natalie Klein
   -

   1 Niklas Wilming
   -

   1 Richard H?chenberger
   -

   1 Sagun Pai
   -

   1 Sourav Singh
   -

   1 Tom Dupr? la Tour
   -

   1 kambysese
   -

   1 pbnsilva
   -

   1 sviter
   -

   1 zuxfoucault
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160511/c756059e/attachment.html>

From bertrand.thirion at inria.fr  Wed May 11 13:56:08 2016
From: bertrand.thirion at inria.fr (Bertrand Thirion)
Date: Wed, 11 May 2016 19:56:08 +0200 (CEST)
Subject: [Neuroimaging] [ANN] MNE-Python 0.12
In-Reply-To: <CADeotZoTuVV+FEmmxbA1cqo7wtyoSTPm=CQvfRZwxVTFN4r=qA@mail.gmail.com>
References: <CADeotZoTuVV+FEmmxbA1cqo7wtyoSTPm=CQvfRZwxVTFN4r=qA@mail.gmail.com>
Message-ID: <276662165.30038423.1462989368907.JavaMail.zimbra@inria.fr>

Congratulations ! 

Bertrand 

----- Mail original -----

> De: "Alexandre Gramfort" <alexandre.gramfort at telecom-paristech.fr>
> ?: "Neuroimaging analysis in Python" <neuroimaging at python.org>, "mne
> analysis" <mne_analysis at nmr.mgh.harvard.edu>,
> neurospin-time-group at googlegroups.com, megcommunity at jiscmail.ac.uk
> Envoy?: Mercredi 11 Mai 2016 09:59:36
> Objet: [Neuroimaging] [ANN] MNE-Python 0.12

> Hi,

> We are pleased to announce the new 0.12 release of MNE-Python. This release
> comes with many improvements to usability, visualization and documentation
> and bug fixes.

> A few highlights:

> *

> We entirely revamped our documentation at the MNE website with a new
> easy-to-follow structure, have a look at http://martinos.org/mne and let us
> know if you would like to read more on a particular topic. See eg.
> http://martinos.org/mne/stable/tutorials.html
> *

> We introduced annotations for marking arbitrary segments of raw data. This
> can be used in order to annotate M/EEG recordings with naturalistic stimuli
> or for rejecting bad segments of data. See
> http://martinos.org/mne/dev/auto_tutorials/plot_brainstorm_auditory.html for
> an example.
> *

> Added the ability to create animations/movies of sensor topographies. See
> Evoked.animate_topomap method.
> *

> We now have movement compensation for Maxwell-Filter
> *

> We have new short-hand plotting function for showing sensor positions and
> layouts.
> *

> We now explicitly support ECoG data with a specific ecog channel type.
> *

> Evoked activity (as a butterfly time series) and corresponding topomaps can
> now be shown in one plot with `Evoked.plot_joint()` for spatio-temporal
> brain dynamics
> *

> Support for reading and estimation of fixed-position dipole time courses
> (similar to Elekta ``xfit``)
> *

> New mne.io.read_raw_cnt function for reading Neuroscan CNT files

> Notable API changes:

> *

> To unify in-place modification vs. copying API, the `copy` parameter was
> deprecated for all MNE object methods and will be removed in a later
> version; instead, `inst.copy().method()` is to be used. Also, all object
> methods now return `self`, allowing reliable chaining (e.g. `raw_resampled =
> raw.copy().filter(1).resample(100)`)
> *

> Generalization Across Time now supports custom predict functions, e.g.
> predicting probabilities rather than classes, via the `predict_method`
> keyword argument; and an option was added to score either across or within
> folds via the `predict_mode` keyword argument.
> *

> We now have additional decimation parameters for time-frequency methods
> *

> When estimating covariance from raw data, the same regularization methods can
> be used as for estimating the covariance from epoched data.
> *

> From now on ECG, EOG and EMG channels are shown by default in butterfly plots

> For a full list of improvements and API changes, see:

> http://martinos.org/mne/stable/whats_new.html#version-0-12

> To install the latest release the following command should do the job:

> pip install --upgrade --user mne

> As usual we welcome your bug reports, feature requests, critiques and

> contributions.

> Some links:

> - https://github.com/mne-tools/mne-python (code + readme on how to install)

> - http://martinos.org/mne/stable/ (full MNE documentation)

> Follow us on Twitter: https://twitter.com/mne_python

> Regards,

> The MNE-Python developers

> People who contributed to this release with their number of commits:

> The committer list for this release is the following (preceded by

> number of commits):

> *

> 348 Eric Larson
> *

> 347 Jaakko Leppakangas
> *

> 157 Alexandre Gramfort
> *

> 139 Jona Sassenhagen
> *

> 67 Jean-Remi King
> *

> 32 Chris Holdgraf
> *

> 31 Denis A. Engemann
> *

> 30 Mainak Jas
> *

> 16 Christopher J. Bailey
> *

> 13 Marijn van Vliet
> *

> 10 Mark Wronkiewicz
> *

> 9 Teon Brooks
> *

> 9 kaichogami
> *

> 8 Cl?ment Moutard
> *

> 5 Camilo Lamus
> *

> 5 mmagnuski
> *

> 4 Christian Brodbeck
> *

> 4 Daniel McCloy
> *

> 4 Yousra Bekhti
> *

> 3 Fede Raimondo
> *

> 1 Jussi Nurminen
> *

> 1 MartinBaBer
> *

> 1 Mikolaj Magnuski
> *

> 1 Natalie Klein
> *

> 1 Niklas Wilming
> *

> 1 Richard H?chenberger
> *

> 1 Sagun Pai
> *

> 1 Sourav Singh
> *

> 1 Tom Dupr? la Tour
> *

> 1 kambysese
> *

> 1 pbnsilva
> *

> 1 sviter
> *

> 1 zuxfoucault

> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160511/0964aa48/attachment.html>

From mfbarton at umich.edu  Sat May 14 22:07:14 2016
From: mfbarton at umich.edu (Michael Barton)
Date: Sat, 14 May 2016 22:07:14 -0400
Subject: [Neuroimaging] pysurfer
Message-ID: <8FEB3F3B-C7F0-4F41-894B-FB63C52E6D46@umich.edu>

Hi,
Is it possible to get pysurfer with python 3.5? Or only 2.7?

Thanks,
Michael

From mwaskom at stanford.edu  Sun May 15 16:25:41 2016
From: mwaskom at stanford.edu (Michael Waskom)
Date: Sun, 15 May 2016 13:25:41 -0700
Subject: [Neuroimaging] pysurfer
In-Reply-To: <8FEB3F3B-C7F0-4F41-894B-FB63C52E6D46@umich.edu>
References: <8FEB3F3B-C7F0-4F41-894B-FB63C52E6D46@umich.edu>
Message-ID: <CAKZ-Uq2d4agfnNvd=5OOnG4=1v5yu-hpktUJcAKQZM=DbFLtRQ@mail.gmail.com>

Hi Michael,

PySurfer currently works only with Python 2.7. The latest release of Mayavi
was the first to support Python 3. Hopefully support will soon be extended
to PySurfer, but someone needs to actually do the work:
https://github.com/nipy/PySurfer/issues/143

Best,
Michael

On Sat, May 14, 2016 at 7:07 PM, Michael Barton <mfbarton at umich.edu> wrote:

> Hi,
> Is it possible to get pysurfer with python 3.5? Or only 2.7?
>
> Thanks,
> Michael
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160515/3448010e/attachment.html>

From larson.eric.d at gmail.com  Mon May 16 14:03:33 2016
From: larson.eric.d at gmail.com (Eric Larson)
Date: Mon, 16 May 2016 14:03:33 -0400
Subject: [Neuroimaging] PySurfer Py3k support
Message-ID: <CAGu2niU-CkRKz2NYhC7E-CiiCjUXeArzF_vhyPdT3Y-g2tFSJQ@mail.gmail.com>

It only required a few changes, feel free to try the PR if you're curious:

https://github.com/nipy/PySurfer/pull/152

Right now it's not easy to use since Mayavi and VTK need to be installed
manually, but the Anaconda folks have some packaging in progress that will
hopefully make it easier soon.

Cheers,
Eric
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160516/c54b1189/attachment.html>

From mahmoud.zeydabadinezhad at emory.edu  Mon May 16 16:08:12 2016
From: mahmoud.zeydabadinezhad at emory.edu (Zeydabadinezhad, Mahmoud)
Date: Mon, 16 May 2016 20:08:12 +0000
Subject: [Neuroimaging] Scrubbing DTI images
Message-ID: <SN2PR05MB249401F489D08B22C7F9ADD0E4770@SN2PR05MB2494.namprd05.prod.outlook.com>

Dear all,

I have some DTI images collected in 61 directions + 6 B0s in 3T. I was wondering if you could give me a hint how can I do scrubbing on my DTI data either using FSL or DIPY or any other script.
By scrubbing, I mean the same idea used in fMRI community i.e. detecting the outlier/motion corrupted slices or volumes and removing them from DTI calculation.
I appreciate your time and help.

Thank you!
Mahmoud


________________________________

This e-mail message (including any attachments) is for the sole use of
the intended recipient(s) and may contain confidential and privileged
information. If the reader of this message is not the intended
recipient, you are hereby notified that any dissemination, distribution
or copying of this message (including any attachments) is strictly
prohibited.

If you have received this message in error, please contact
the sender by reply e-mail message and destroy all copies of the
original message (including attachments).
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160516/87511276/attachment.html>

From krzysztof.gorgolewski at gmail.com  Tue May 17 16:34:52 2016
From: krzysztof.gorgolewski at gmail.com (Chris Filo Gorgolewski)
Date: Tue, 17 May 2016 13:34:52 -0700
Subject: [Neuroimaging] Scrubbing DTI images
In-Reply-To: <SN2PR05MB249401F489D08B22C7F9ADD0E4770@SN2PR05MB2494.namprd05.prod.outlook.com>
References: <SN2PR05MB249401F489D08B22C7F9ADD0E4770@SN2PR05MB2494.namprd05.prod.outlook.com>
Message-ID: <CAAQzouO56wwOYB6S=giZsikUTnSwALdS0kqmz2Tbo=63Fst5wA@mail.gmail.com>

You should look into RESTORE which fits DTI with outliers removed/down
weighted: http://nipy.org/dipy/examples_built/restore_dti.html

On Mon, May 16, 2016 at 1:08 PM, Zeydabadinezhad, Mahmoud <
mahmoud.zeydabadinezhad at emory.edu> wrote:

> Dear all,
>
> I have some DTI images collected in 61 directions + 6 B0s in 3T. I was
> wondering if you could give me a hint how can I do scrubbing on my DTI data
> either using FSL or DIPY or any other script.
> By scrubbing, I mean the same idea used in fMRI community i.e. detecting
> the outlier/motion corrupted slices or volumes and removing them from DTI
> calculation.
> I appreciate your time and help.
>
> Thank you!
> Mahmoud
>
>
> ------------------------------
>
> This e-mail message (including any attachments) is for the sole use of
> the intended recipient(s) and may contain confidential and privileged
> information. If the reader of this message is not the intended
> recipient, you are hereby notified that any dissemination, distribution
> or copying of this message (including any attachments) is strictly
> prohibited.
>
> If you have received this message in error, please contact
> the sender by reply e-mail message and destroy all copies of the
> original message (including attachments).
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160517/1fd5df19/attachment.html>

From code at oscaresteban.es  Tue May 17 16:43:12 2016
From: code at oscaresteban.es (Oscar Esteban)
Date: Tue, 17 May 2016 13:43:12 -0700
Subject: [Neuroimaging] Scrubbing DTI images
In-Reply-To: <CAAQzouO56wwOYB6S=giZsikUTnSwALdS0kqmz2Tbo=63Fst5wA@mail.gmail.com>
References: <SN2PR05MB249401F489D08B22C7F9ADD0E4770@SN2PR05MB2494.namprd05.prod.outlook.com>
 <CAAQzouO56wwOYB6S=giZsikUTnSwALdS0kqmz2Tbo=63Fst5wA@mail.gmail.com>
Message-ID: <CAAsFcgw8LKUQjxM+un_wugbqUdMbX59jyrt_5YHebL5VBvGw1A@mail.gmail.com>

Hi Mahmoud,

Just to add some to Chris' response. RESTORE will fit the DTI rejecting
outliers "voxel-wise".

Probably you want first to run something like FSL Eddy (
http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/EDDY/UsersGuide). Eddy will deal with
head motion, eddy currents and susceptibility distortions. One approach to
remove outliers is denoising, i.e. using
http://nipy.org/dipy/examples_built/denoise_nlmeans.html

If you do not denoise, RESTORE may a good option if you want to fit DTI.

Cheers,
Oscar

On Tue, May 17, 2016 at 1:34 PM, Chris Filo Gorgolewski <
krzysztof.gorgolewski at gmail.com> wrote:

> You should look into RESTORE which fits DTI with outliers removed/down
> weighted: http://nipy.org/dipy/examples_built/restore_dti.html
>
> On Mon, May 16, 2016 at 1:08 PM, Zeydabadinezhad, Mahmoud <
> mahmoud.zeydabadinezhad at emory.edu> wrote:
>
>> Dear all,
>>
>> I have some DTI images collected in 61 directions + 6 B0s in 3T. I was
>> wondering if you could give me a hint how can I do scrubbing on my DTI data
>> either using FSL or DIPY or any other script.
>> By scrubbing, I mean the same idea used in fMRI community i.e. detecting
>> the outlier/motion corrupted slices or volumes and removing them from DTI
>> calculation.
>> I appreciate your time and help.
>>
>> Thank you!
>> Mahmoud
>>
>>
>> ------------------------------
>>
>> This e-mail message (including any attachments) is for the sole use of
>> the intended recipient(s) and may contain confidential and privileged
>> information. If the reader of this message is not the intended
>> recipient, you are hereby notified that any dissemination, distribution
>> or copying of this message (including any attachments) is strictly
>> prohibited.
>>
>> If you have received this message in error, please contact
>> the sender by reply e-mail message and destroy all copies of the
>> original message (including attachments).
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160517/126755da/attachment.html>

From garyfallidis at gmail.com  Tue May 17 18:16:05 2016
From: garyfallidis at gmail.com (Eleftherios Garyfallidis)
Date: Tue, 17 May 2016 22:16:05 +0000
Subject: [Neuroimaging] Scrubbing DTI images
In-Reply-To: <CAAsFcgw8LKUQjxM+un_wugbqUdMbX59jyrt_5YHebL5VBvGw1A@mail.gmail.com>
References: <SN2PR05MB249401F489D08B22C7F9ADD0E4770@SN2PR05MB2494.namprd05.prod.outlook.com>
 <CAAQzouO56wwOYB6S=giZsikUTnSwALdS0kqmz2Tbo=63Fst5wA@mail.gmail.com>
 <CAAsFcgw8LKUQjxM+un_wugbqUdMbX59jyrt_5YHebL5VBvGw1A@mail.gmail.com>
Message-ID: <CAN6JmWP17oybTZ4Zky=EF1Qc8noS+CT=d=8FjcvHptt0jgRC4Q@mail.gmail.com>

Hi Mahmoud,

What you need to do using RESTORE is find the dwi volumes which are
outliers (using the fit errors) and remove them from the data.

Does anyone has written a function for doing this automatically? If yes
please be happy to share it with the rest of us here and of course with
Mahmoud. Otherwise, Mahmoud you can try to do this by yourself by looking
at the tutorial given above but for this you will have to put some extra
effort to understand the different DIPY classes and functions. We are here
to help, so let us know how it goes.

Best,
Eleftherios



On Tue, May 17, 2016 at 5:09 PM Oscar Esteban <code at oscaresteban.es> wrote:

> Hi Mahmoud,
>
> Just to add some to Chris' response. RESTORE will fit the DTI rejecting
> outliers "voxel-wise".
>
> Probably you want first to run something like FSL Eddy (
> http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/EDDY/UsersGuide). Eddy will deal
> with head motion, eddy currents and susceptibility distortions. One
> approach to remove outliers is denoising, i.e. using
> http://nipy.org/dipy/examples_built/denoise_nlmeans.html
>
> If you do not denoise, RESTORE may a good option if you want to fit DTI.
>
> Cheers,
> Oscar
>
> On Tue, May 17, 2016 at 1:34 PM, Chris Filo Gorgolewski <
> krzysztof.gorgolewski at gmail.com> wrote:
>
>> You should look into RESTORE which fits DTI with outliers removed/down
>> weighted: http://nipy.org/dipy/examples_built/restore_dti.html
>>
>> On Mon, May 16, 2016 at 1:08 PM, Zeydabadinezhad, Mahmoud <
>> mahmoud.zeydabadinezhad at emory.edu> wrote:
>>
>>> Dear all,
>>>
>>> I have some DTI images collected in 61 directions + 6 B0s in 3T. I was
>>> wondering if you could give me a hint how can I do scrubbing on my DTI data
>>> either using FSL or DIPY or any other script.
>>> By scrubbing, I mean the same idea used in fMRI community i.e. detecting
>>> the outlier/motion corrupted slices or volumes and removing them from DTI
>>> calculation.
>>> I appreciate your time and help.
>>>
>>> Thank you!
>>> Mahmoud
>>>
>>>
>>> ------------------------------
>>>
>>> This e-mail message (including any attachments) is for the sole use of
>>> the intended recipient(s) and may contain confidential and privileged
>>> information. If the reader of this message is not the intended
>>> recipient, you are hereby notified that any dissemination, distribution
>>> or copying of this message (including any attachments) is strictly
>>> prohibited.
>>>
>>> If you have received this message in error, please contact
>>> the sender by reply e-mail message and destroy all copies of the
>>> original message (including attachments).
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160517/25e2a426/attachment.html>

From zeydabadi at gmail.com  Tue May 17 20:46:21 2016
From: zeydabadi at gmail.com (Mahmoud)
Date: Tue, 17 May 2016 20:46:21 -0400
Subject: [Neuroimaging] Scrubbing DTI images
In-Reply-To: <CAAsFcgw8LKUQjxM+un_wugbqUdMbX59jyrt_5YHebL5VBvGw1A@mail.gmail.com>
References: <SN2PR05MB249401F489D08B22C7F9ADD0E4770@SN2PR05MB2494.namprd05.prod.outlook.com>
 <CAAQzouO56wwOYB6S=giZsikUTnSwALdS0kqmz2Tbo=63Fst5wA@mail.gmail.com>
 <CAAsFcgw8LKUQjxM+un_wugbqUdMbX59jyrt_5YHebL5VBvGw1A@mail.gmail.com>
Message-ID: <CAHBa8XRUM5VxOhGVPXDCfbDzDNSY0GqiwEwp_BXrC3WuoQKLSw@mail.gmail.com>

Thank you all for the hints.

Oscar,
Could you elaborate more about what you suggested? I mean what exactly
denoising is doing?

Thank you!
Mahmoud

On Tue, May 17, 2016 at 4:43 PM, Oscar Esteban <code at oscaresteban.es> wrote:

> Hi Mahmoud,
>
> Just to add some to Chris' response. RESTORE will fit the DTI rejecting
> outliers "voxel-wise".
>
> Probably you want first to run something like FSL Eddy (
> http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/EDDY/UsersGuide). Eddy will deal
> with head motion, eddy currents and susceptibility distortions. One
> approach to remove outliers is denoising, i.e. using
> http://nipy.org/dipy/examples_built/denoise_nlmeans.html
>
> If you do not denoise, RESTORE may a good option if you want to fit DTI.
>
> Cheers,
> Oscar
>
> On Tue, May 17, 2016 at 1:34 PM, Chris Filo Gorgolewski <
> krzysztof.gorgolewski at gmail.com> wrote:
>
>> You should look into RESTORE which fits DTI with outliers removed/down
>> weighted: http://nipy.org/dipy/examples_built/restore_dti.html
>>
>> On Mon, May 16, 2016 at 1:08 PM, Zeydabadinezhad, Mahmoud <
>> mahmoud.zeydabadinezhad at emory.edu> wrote:
>>
>>> Dear all,
>>>
>>> I have some DTI images collected in 61 directions + 6 B0s in 3T. I was
>>> wondering if you could give me a hint how can I do scrubbing on my DTI data
>>> either using FSL or DIPY or any other script.
>>> By scrubbing, I mean the same idea used in fMRI community i.e. detecting
>>> the outlier/motion corrupted slices or volumes and removing them from DTI
>>> calculation.
>>> I appreciate your time and help.
>>>
>>> Thank you!
>>> Mahmoud
>>>
>>>
>>> ------------------------------
>>>
>>> This e-mail message (including any attachments) is for the sole use of
>>> the intended recipient(s) and may contain confidential and privileged
>>> information. If the reader of this message is not the intended
>>> recipient, you are hereby notified that any dissemination, distribution
>>> or copying of this message (including any attachments) is strictly
>>> prohibited.
>>>
>>> If you have received this message in error, please contact
>>> the sender by reply e-mail message and destroy all copies of the
>>> original message (including attachments).
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160517/92e7026a/attachment.html>

From stjeansam at gmail.com  Tue May 17 21:03:13 2016
From: stjeansam at gmail.com (Samuel St-Jean)
Date: Wed, 18 May 2016 03:03:13 +0200
Subject: [Neuroimaging] Scrubbing DTI images
In-Reply-To: <CAHBa8XRUM5VxOhGVPXDCfbDzDNSY0GqiwEwp_BXrC3WuoQKLSw@mail.gmail.com>
References: <SN2PR05MB249401F489D08B22C7F9ADD0E4770@SN2PR05MB2494.namprd05.prod.outlook.com>
 <CAAQzouO56wwOYB6S=giZsikUTnSwALdS0kqmz2Tbo=63Fst5wA@mail.gmail.com>
 <CAAsFcgw8LKUQjxM+un_wugbqUdMbX59jyrt_5YHebL5VBvGw1A@mail.gmail.com>
 <CAHBa8XRUM5VxOhGVPXDCfbDzDNSY0GqiwEwp_BXrC3WuoQKLSw@mail.gmail.com>
Message-ID: <CAKADGuoXiD9OX5vRChOX6m0bbiBw+QJY8S=Xas1kwMv_Upb6SQ@mail.gmail.com>

Denoising will give you back plausible values based on neighborhood
information, it will not do whole volume rejection, restore will do that
for you though based on the tensor model.
On May 18, 2016 8:53 AM, "Mahmoud" <zeydabadi at gmail.com> wrote:

> Thank you all for the hints.
>
> Oscar,
> Could you elaborate more about what you suggested? I mean what exactly
> denoising is doing?
>
> Thank you!
> Mahmoud
>
> On Tue, May 17, 2016 at 4:43 PM, Oscar Esteban <code at oscaresteban.es>
> wrote:
>
>> Hi Mahmoud,
>>
>> Just to add some to Chris' response. RESTORE will fit the DTI rejecting
>> outliers "voxel-wise".
>>
>> Probably you want first to run something like FSL Eddy (
>> http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/EDDY/UsersGuide). Eddy will deal
>> with head motion, eddy currents and susceptibility distortions. One
>> approach to remove outliers is denoising, i.e. using
>> http://nipy.org/dipy/examples_built/denoise_nlmeans.html
>>
>> If you do not denoise, RESTORE may a good option if you want to fit DTI.
>>
>> Cheers,
>> Oscar
>>
>> On Tue, May 17, 2016 at 1:34 PM, Chris Filo Gorgolewski <
>> krzysztof.gorgolewski at gmail.com> wrote:
>>
>>> You should look into RESTORE which fits DTI with outliers removed/down
>>> weighted: http://nipy.org/dipy/examples_built/restore_dti.html
>>>
>>> On Mon, May 16, 2016 at 1:08 PM, Zeydabadinezhad, Mahmoud <
>>> mahmoud.zeydabadinezhad at emory.edu> wrote:
>>>
>>>> Dear all,
>>>>
>>>> I have some DTI images collected in 61 directions + 6 B0s in 3T. I was
>>>> wondering if you could give me a hint how can I do scrubbing on my DTI data
>>>> either using FSL or DIPY or any other script.
>>>> By scrubbing, I mean the same idea used in fMRI community i.e.
>>>> detecting the outlier/motion corrupted slices or volumes and removing them
>>>> from DTI calculation.
>>>> I appreciate your time and help.
>>>>
>>>> Thank you!
>>>> Mahmoud
>>>>
>>>>
>>>> ------------------------------
>>>>
>>>> This e-mail message (including any attachments) is for the sole use of
>>>> the intended recipient(s) and may contain confidential and privileged
>>>> information. If the reader of this message is not the intended
>>>> recipient, you are hereby notified that any dissemination, distribution
>>>> or copying of this message (including any attachments) is strictly
>>>> prohibited.
>>>>
>>>> If you have received this message in error, please contact
>>>> the sender by reply e-mail message and destroy all copies of the
>>>> original message (including attachments).
>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>>
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160518/c480b7e2/attachment.html>

From vivekjoshi1894 at gmail.com  Wed May 18 01:33:11 2016
From: vivekjoshi1894 at gmail.com (Vivek Joshi)
Date: Wed, 18 May 2016 11:03:11 +0530
Subject: [Neuroimaging] Regarding acquisition of more 3d dwi images
Message-ID: <573bfe9a.8715430a.85a98.5e63@mx.google.com>

Hello all
There is a need for more 3d images like sherbrooke_3shell (HARDI datasets) so that a demonstration can be made on more images.
I got stanford_hardi dataset through another code but i need atleast 3 more datasets representing brain 3d mri images.
So it would be of great help if you send us any LINK where we can download the images.
Please help!!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160518/93bc2dfb/attachment.html>

From stjeansam at gmail.com  Wed May 18 03:07:46 2016
From: stjeansam at gmail.com (Samuel St-Jean)
Date: Wed, 18 May 2016 09:07:46 +0200
Subject: [Neuroimaging] Regarding acquisition of more 3d dwi images
In-Reply-To: <573bfe9a.8715430a.85a98.5e63@mx.google.com>
References: <573bfe9a.8715430a.85a98.5e63@mx.google.com>
Message-ID: <CAKADGuqfUsQSStozX4Bgwe9jV1v0URahsgxY8kPF+B_spt6mLw@mail.gmail.com>

Try the hcp datasets, all same acquisitions, high res, multishell, over 500
subjects and freely available.

For other datasets, like the sherbrooke one, its only people putting it
online for others to re-use. I also have a 1.2mm and 1.9mm from the same
subject (but probably different from the 3 shell one, scanned in Sherbrooke
also) over here if you want https://github.com/samuelstjean/nlsam_data
On May 18, 2016 14:35, "Vivek Joshi" <vivekjoshi1894 at gmail.com> wrote:

> Hello all
> There is a need for more 3d images like sherbrooke_3shell (HARDI datasets)
> so that a demonstration can be made on more images.
> I got stanford_hardi dataset through another code but i need atleast 3
> more datasets representing brain 3d mri images.
> So it would be of great help if you send us any LINK where we can download
> the images.
> Please help!!
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160518/967c1e0f/attachment.html>

From arokem at gmail.com  Wed May 18 10:15:52 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Wed, 18 May 2016 07:15:52 -0700
Subject: [Neuroimaging] Regarding acquisition of more 3d dwi images
In-Reply-To: <CAKADGuqfUsQSStozX4Bgwe9jV1v0URahsgxY8kPF+B_spt6mLw@mail.gmail.com>
References: <573bfe9a.8715430a.85a98.5e63@mx.google.com>
 <CAKADGuqfUsQSStozX4Bgwe9jV1v0URahsgxY8kPF+B_spt6mLw@mail.gmail.com>
Message-ID: <CA+X4w0wsPYVbo1qZBm1WERSYFriQ-OG9EdHSShDVDDasRK1VAg@mail.gmail.com>

Hi Vivek,

On Wed, May 18, 2016 at 12:07 AM, Samuel St-Jean <stjeansam at gmail.com>
wrote:

> Try the hcp datasets, all same acquisitions, high res, multishell, over
> 500 subjects and freely available.
>
> For other datasets, like the sherbrooke one, its only people putting it
> online for others to re-use. I also have a 1.2mm and 1.9mm from the same
> subject (but probably different from the 3 shell one, scanned in Sherbrooke
> also) over here if you want https://github.com/samuelstjean/nlsam_data
>


Another good place to look is NITRC: https://www.nitrc.org/ -- there are
several DWI datasets available to download directly, and links to several
more that can be downloaded elsewhere.

Best,

Ariel



> On May 18, 2016 14:35, "Vivek Joshi" <vivekjoshi1894 at gmail.com> wrote:
>
>> Hello all
>> There is a need for more 3d images like sherbrooke_3shell (HARDI
>> datasets) so that a demonstration can be made on more images.
>> I got stanford_hardi dataset through another code but i need atleast 3
>> more datasets representing brain 3d mri images.
>> So it would be of great help if you send us any LINK where we can
>> download the images.
>> Please help!!
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160518/7b8f2f87/attachment.html>

From jhlegarreta at vicomtech.org  Wed May 18 13:10:39 2016
From: jhlegarreta at vicomtech.org (Jon Haitz Legarreta)
Date: Wed, 18 May 2016 19:10:39 +0200
Subject: [Neuroimaging] [dipy] Issues trying to install dipy
Message-ID: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>

Hi there,
I'm a newbie to dipy.

I was trying to follow the instructions in [1] to have dipy installed from
the source code, so that I could execute the dipy examples.

I'm using Windows 10 and Anaconda 3.

When trying to execute
*python setup.py develop*

the Anaconda prompt yields an error that says in the end:


*File: "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126,
__init__if self.ld_version >= "2.10.90":TypeError: unorderable types:
NoneType() >= str()*

I've been googling for a solution without success.

I don't know whether this looks like Anaconda3 is trying to use cygwin
instead of mingw32, and whether that is the root cause.

In either case, does anyone know how to solve the issue?

Attached is the trace (it's short) of the error if this is of any help.


Thank you,
JON HAITZ


[1] http://nipy.org/dipy/installation.html#install-source-nix
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160518/744c055e/attachment.html>
-------------- next part --------------
C:\Anaconda3\lib\site-packages\setuptools-19.6.2-py3.5.egg\setuptools\dist.py:285: UserWarning: Normalizing '0.12.0dev' to '0.12.0.dev0'
running develop
running egg_info
writing dipy.egg-info\PKG-INFO
writing top-level names to dipy.egg-info\top_level.txt
package init file 'dipy\data\tests\__init__.py' not found (or not a regular file)
writing dependency_links to dipy.egg-info\dependency_links.txt
writing requirements to dipy.egg-info\requires.txt
reading manifest file 'dipy.egg-info\SOURCES.txt'
reading manifest template 'MANIFEST.in'
warning: no files found matching 'TODO'
writing manifest file 'dipy.egg-info\SOURCES.txt'
running build_ext
Traceback (most recent call last):
  File "setup.py", line 227, in <module>
    main(**extra_setuptools_args)
  File "setup.py", line 220, in main
    **extra_args
  File "C:\Anaconda3\lib\distutils\core.py", line 148, in setup
    dist.run_commands()
  File "C:\Anaconda3\lib\distutils\dist.py", line 955, in run_commands
    self.run_command(cmd)
  File "C:\Anaconda3\lib\distutils\dist.py", line 974, in run_command
    cmd_obj.run()
  File "C:\Anaconda3\lib\site-packages\setuptools-19.6.2-py3.5.egg\setuptools\command\develop.py", line 34, in run
  File "C:\Anaconda3\lib\site-packages\setuptools-19.6.2-py3.5.egg\setuptools\command\develop.py", line 119, in install_for_development
  File "C:\Anaconda3\lib\distutils\cmd.py", line 313, in run_command
    self.distribution.run_command(command)
  File "C:\Anaconda3\lib\distutils\dist.py", line 974, in run_command
    cmd_obj.run()
  File "C:\Anaconda3\lib\site-packages\Cython\Distutils\build_ext.py", line 164, in run
    _build_ext.build_ext.run(self)
  File "C:\Anaconda3\lib\distutils\command\build_ext.py", line 307, in run
    force=self.force)
  File "C:\Anaconda3\lib\distutils\ccompiler.py", line 1031, in new_compiler
    return klass(None, dry_run, force)
  File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 282, in __init__
    CygwinCCompiler.__init__ (self, verbose, dry_run, force)
  File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126, in __init__
    if self.ld_version >= "2.10.90":
TypeError: unorderable types: NoneType() >= str()

From vivekjoshi1894 at gmail.com  Wed May 18 14:07:51 2016
From: vivekjoshi1894 at gmail.com (Vivek Joshi)
Date: Wed, 18 May 2016 23:37:51 +0530
Subject: [Neuroimaging] cannot import the 3d images
Message-ID: <CAMYkJfv53Mh0yt-tjMnkTVrcmSxdWY1UFA+PwnoY6m2u-8QTQg@mail.gmail.com>

Thank you Samuel St-Jean and Ariel Rokem!
But the image datasets you mentioned cannot be imported in the PIESNO code.
Only stanford_hardi and sherbrooke_3shell are fetched properly.
So please suggest a way to run the code without any error.
THANK YOU!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160518/fdf862f8/attachment.html>

From arokem at gmail.com  Wed May 18 14:19:01 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Wed, 18 May 2016 11:19:01 -0700
Subject: [Neuroimaging] cannot import the 3d images
In-Reply-To: <CAMYkJfv53Mh0yt-tjMnkTVrcmSxdWY1UFA+PwnoY6m2u-8QTQg@mail.gmail.com>
References: <CAMYkJfv53Mh0yt-tjMnkTVrcmSxdWY1UFA+PwnoY6m2u-8QTQg@mail.gmail.com>
Message-ID: <CA+X4w0zdCpcwW42+Bk6++pTMNv9w6+y5tOm8T2DmGH1m_78U5w@mail.gmail.com>

Hi Vivek,

On Wed, May 18, 2016 at 11:07 AM, Vivek Joshi <vivekjoshi1894 at gmail.com>
wrote:

> Thank you Samuel St-Jean and Ariel Rokem!
> But the image datasets you mentioned cannot be imported in the PIESNO
> code. Only stanford_hardi and sherbrooke_3shell are fetched properly.
> So please suggest a way to run the code without any error.
> THANK YOU!
>

You are correct that Dipy itself does not include easy-to-use data-fetchers
for many of these data-sets (though there are in fact a couple more in the
dipy.data module: http://nipy.org/dipy/reference/dipy.data.html). You
should download the data files to your computer and use the nibabel library
(http://nipy.org/nibabel/) to read the data from the files.

Best,

Ariel




> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160518/bd08b0ef/attachment.html>

From garyfallidis at gmail.com  Wed May 18 17:09:04 2016
From: garyfallidis at gmail.com (Eleftherios Garyfallidis)
Date: Wed, 18 May 2016 21:09:04 +0000
Subject: [Neuroimaging] cannot import the 3d images
In-Reply-To: <CA+X4w0zdCpcwW42+Bk6++pTMNv9w6+y5tOm8T2DmGH1m_78U5w@mail.gmail.com>
References: <CAMYkJfv53Mh0yt-tjMnkTVrcmSxdWY1UFA+PwnoY6m2u-8QTQg@mail.gmail.com>
 <CA+X4w0zdCpcwW42+Bk6++pTMNv9w6+y5tOm8T2DmGH1m_78U5w@mail.gmail.com>
Message-ID: <CAN6JmWOEfL5mX_FiXUMwk+wP5U23jnD3o72a6YKUg3EGQXqSsg@mail.gmail.com>

Please have a look at the quick start

http://nipy.org/dipy/examples_built/quick_start.html#example-quick-start
Any Nifti image can be loaded if you know the filepath in the disk

fdwi = /home/username/some_director/dwi.nii.gz
import nibabel as nib
img = nib.load(fdwi)
data = img.get_data()

Keep it up!

On Wed, May 18, 2016 at 2:19 PM Ariel Rokem <arokem at gmail.com> wrote:

> Hi Vivek,
>
> On Wed, May 18, 2016 at 11:07 AM, Vivek Joshi <vivekjoshi1894 at gmail.com>
> wrote:
>
>> Thank you Samuel St-Jean and Ariel Rokem!
>> But the image datasets you mentioned cannot be imported in the PIESNO
>> code. Only stanford_hardi and sherbrooke_3shell are fetched properly.
>> So please suggest a way to run the code without any error.
>> THANK YOU!
>>
>
> You are correct that Dipy itself does not include easy-to-use
> data-fetchers for many of these data-sets (though there are in fact a
> couple more in the dipy.data module:
> http://nipy.org/dipy/reference/dipy.data.html). You should download the
> data files to your computer and use the nibabel library (
> http://nipy.org/nibabel/) to read the data from the files.
>
> Best,
>
> Ariel
>
>
>
>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160518/743dd732/attachment.html>

From sulantha.s at gmail.com  Thu May 19 13:50:26 2016
From: sulantha.s at gmail.com (Sulantha Sanjeewa)
Date: Thu, 19 May 2016 13:50:26 -0400
Subject: [Neuroimaging] [dipy] Eddy current correction
Message-ID: <CAPM2tnhmvJ+giEt+dxbE_9Msgc+LYfLe_qacgnv2JpqB=+6EXw@mail.gmail.com>

Hi everyone,
I was wondering if the dipy package includes functions to do eddy current
correction. I was going through the 2014 publication, and it mentions plans
to implement this.
Can you please let me know if this is available, and if not an alternative?
Best regards,
Sulantha.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160519/369f0310/attachment.html>

From vivekjoshi1894 at gmail.com  Fri May 20 09:56:05 2016
From: vivekjoshi1894 at gmail.com (Vivek Joshi)
Date: Fri, 20 May 2016 19:26:05 +0530
Subject: [Neuroimaging] attribute error
Message-ID: <CAMYkJfuKjAbAHcMqPAdCeUEEVasOiEHUEWsfa0SL-vbh1_phnw@mail.gmail.com>

Thank you all once again!
I downloaded some of the dipy.data examples. Some of them got fetched
easily like taiwan_ntu_dsi.
But when i tried to fetch others it showed an attribute error ('tuple'
object has no attribute).
So can you please help me how to fetch it the PIESNO code.

Thank You
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160520/0b9c6d3e/attachment.html>

From arokem at gmail.com  Fri May 20 10:33:16 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Fri, 20 May 2016 07:33:16 -0700
Subject: [Neuroimaging] attribute error
In-Reply-To: <CAMYkJfuKjAbAHcMqPAdCeUEEVasOiEHUEWsfa0SL-vbh1_phnw@mail.gmail.com>
References: <CAMYkJfuKjAbAHcMqPAdCeUEEVasOiEHUEWsfa0SL-vbh1_phnw@mail.gmail.com>
Message-ID: <CA+X4w0ycU_oR+jaz3gjkiS8hkpM7USRoSuQexmmthPkr-1MzqQ@mail.gmail.com>

Hi Vivek,

On Fri, May 20, 2016 at 6:56 AM, Vivek Joshi <vivekjoshi1894 at gmail.com>
wrote:

> Thank you all once again!
> I downloaded some of the dipy.data examples. Some of them got fetched
> easily like taiwan_ntu_dsi.
> But when i tried to fetch others it showed an attribute error ('tuple'
> object has no attribute).
> So can you please help me how to fetch it the PIESNO code.
>
>
Could you please post the code you ran that raised that error?

Thanks!

Ariel


> Thank You
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160520/5f7e6e38/attachment.html>

From arokem at gmail.com  Fri May 20 10:36:43 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Fri, 20 May 2016 07:36:43 -0700
Subject: [Neuroimaging] [dipy] Eddy current correction
In-Reply-To: <CAPM2tnhmvJ+giEt+dxbE_9Msgc+LYfLe_qacgnv2JpqB=+6EXw@mail.gmail.com>
References: <CAPM2tnhmvJ+giEt+dxbE_9Msgc+LYfLe_qacgnv2JpqB=+6EXw@mail.gmail.com>
Message-ID: <CA+X4w0z4bWv8Sp+bGL9PdqtqGzegV57JPcqzrXLyg7TQxpsNkw@mail.gmail.com>

Hi Sulantha,

On Thu, May 19, 2016 at 10:50 AM, Sulantha Sanjeewa <sulantha.s at gmail.com>
wrote:

> Hi everyone,
> I was wondering if the dipy package includes functions to do eddy current
> correction. I was going through the 2014 publication, and it mentions plans
> to implement this.
> Can you please let me know if this is available, and if not an alternative?
>

No eddy current correction in Dipy yet, although there has been a bit of
progress in recent months (on this branch:
https://github.com/sahmed95/dipy/tree/eddy), and we might see some more
progress moving forward (the author of that branch is one of our GSoC
students this summer). For now, other software packages (e.g., FSL,
http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/EDDY) might serve you well.

Cheers,

Ariel


> Best regards,
> Sulantha.
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160520/fe223773/attachment.html>

From vivekjoshi1894 at gmail.com  Sat May 21 02:36:01 2016
From: vivekjoshi1894 at gmail.com (Vivek Joshi)
Date: Sat, 21 May 2016 12:06:01 +0530
Subject: [Neuroimaging] error code
Message-ID: <CAMYkJft5vWhhR4gRKjss3baj6vT3ABPtnoHnSGN0+Dr662RYmA@mail.gmail.com>

Greetings Ariel Rokem
The following code raised the error
import nibabel as nib
import numpy as np
from dipy.denoise.noise_estimate import piesno
from dipy.data import fetch_isbi2013_2shell, read_isbi2013_2shell


fetch_isbi2013_2shell()
img, gtab = read_isbi2013_2shell()
data = img.get_data()

attribute error here!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160521/ac8f9934/attachment.html>

From vivekjoshi1894 at gmail.com  Sat May 21 02:39:03 2016
From: vivekjoshi1894 at gmail.com (Vivek Joshi)
Date: Sat, 21 May 2016 12:09:03 +0530
Subject: [Neuroimaging] plotting 3d dwi images
Message-ID: <CAMYkJftih-dMcKNi2uh=Ao+qFYmkNBoD+iHkFRx8mtbF=4RMLw@mail.gmail.com>

Is there a way to plot a histogram of an original 3d image like
sherbrooke_3shell and its denoised image so that those two can be compared
and denoising can be witnessed.

Thank You!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160521/c831714c/attachment.html>

From jhlegarreta at vicomtech.org  Sat May 21 09:18:49 2016
From: jhlegarreta at vicomtech.org (Jon Haitz Legarreta)
Date: Sat, 21 May 2016 15:18:49 +0200
Subject: [Neuroimaging] Fwd: [dipy] Issues trying to install dipy
In-Reply-To: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>
References: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>
Message-ID: <CAAKhGh9-emozuiqQv2KpY8hghks-1b-bid1PQufJ7YUwDffoDQ@mail.gmail.com>

Hi there,
has anybody experienced the issue below?

Thanks,
JON HAITZ




---------- Forwarded message ----------
From: Jon Haitz Legarreta <jhlegarreta at vicomtech.org>
Date: 18 May 2016 at 19:10
Subject: [dipy] Issues trying to install dipy
To: neuroimaging at python.org


Hi there,
I'm a newbie to dipy.

I was trying to follow the instructions in [1] to have dipy installed from
the source code, so that I could execute the dipy examples.

I'm using Windows 10 and Anaconda 3.

When trying to execute
*python setup.py develop*

the Anaconda prompt yields an error that says in the end:


*File: "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126,
__init__if self.ld_version >= "2.10.90":TypeError: unorderable types:
NoneType() >= str()*

I've been googling for a solution without success.

I don't know whether this looks like Anaconda3 is trying to use cygwin
instead of mingw32, and whether that is the root cause.

In either case, does anyone know how to solve the issue?

Attached is the trace (it's short) of the error if this is of any help.


Thank you,
JON HAITZ


[1] http://nipy.org/dipy/installation.html#install-source-nix
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160521/4a4fbb3d/attachment.html>
-------------- next part --------------
C:\Anaconda3\lib\site-packages\setuptools-19.6.2-py3.5.egg\setuptools\dist.py:285: UserWarning: Normalizing '0.12.0dev' to '0.12.0.dev0'
running develop
running egg_info
writing dipy.egg-info\PKG-INFO
writing top-level names to dipy.egg-info\top_level.txt
package init file 'dipy\data\tests\__init__.py' not found (or not a regular file)
writing dependency_links to dipy.egg-info\dependency_links.txt
writing requirements to dipy.egg-info\requires.txt
reading manifest file 'dipy.egg-info\SOURCES.txt'
reading manifest template 'MANIFEST.in'
warning: no files found matching 'TODO'
writing manifest file 'dipy.egg-info\SOURCES.txt'
running build_ext
Traceback (most recent call last):
  File "setup.py", line 227, in <module>
    main(**extra_setuptools_args)
  File "setup.py", line 220, in main
    **extra_args
  File "C:\Anaconda3\lib\distutils\core.py", line 148, in setup
    dist.run_commands()
  File "C:\Anaconda3\lib\distutils\dist.py", line 955, in run_commands
    self.run_command(cmd)
  File "C:\Anaconda3\lib\distutils\dist.py", line 974, in run_command
    cmd_obj.run()
  File "C:\Anaconda3\lib\site-packages\setuptools-19.6.2-py3.5.egg\setuptools\command\develop.py", line 34, in run
  File "C:\Anaconda3\lib\site-packages\setuptools-19.6.2-py3.5.egg\setuptools\command\develop.py", line 119, in install_for_development
  File "C:\Anaconda3\lib\distutils\cmd.py", line 313, in run_command
    self.distribution.run_command(command)
  File "C:\Anaconda3\lib\distutils\dist.py", line 974, in run_command
    cmd_obj.run()
  File "C:\Anaconda3\lib\site-packages\Cython\Distutils\build_ext.py", line 164, in run
    _build_ext.build_ext.run(self)
  File "C:\Anaconda3\lib\distutils\command\build_ext.py", line 307, in run
    force=self.force)
  File "C:\Anaconda3\lib\distutils\ccompiler.py", line 1031, in new_compiler
    return klass(None, dry_run, force)
  File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 282, in __init__
    CygwinCCompiler.__init__ (self, verbose, dry_run, force)
  File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126, in __init__
    if self.ld_version >= "2.10.90":
TypeError: unorderable types: NoneType() >= str()

From matthew.brett at gmail.com  Sat May 21 10:30:37 2016
From: matthew.brett at gmail.com (Matthew Brett)
Date: Sat, 21 May 2016 10:30:37 -0400
Subject: [Neuroimaging] Fwd: [dipy] Issues trying to install dipy
In-Reply-To: <CAAKhGh9-emozuiqQv2KpY8hghks-1b-bid1PQufJ7YUwDffoDQ@mail.gmail.com>
References: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>
 <CAAKhGh9-emozuiqQv2KpY8hghks-1b-bid1PQufJ7YUwDffoDQ@mail.gmail.com>
Message-ID: <CAH6Pt5rVree2gyERp3DSXVmC_dS4j4gQ66YJpouO62W4JpvX=g@mail.gmail.com>

Hi,

On Sat, May 21, 2016 at 9:18 AM, Jon Haitz Legarreta
<jhlegarreta at vicomtech.org> wrote:
> Hi there,
> has anybody experienced the issue below?
>
> Thanks,
> JON HAITZ
>
>
>
>
> ---------- Forwarded message ----------
> From: Jon Haitz Legarreta <jhlegarreta at vicomtech.org>
> Date: 18 May 2016 at 19:10
> Subject: [dipy] Issues trying to install dipy
> To: neuroimaging at python.org
>
>
> Hi there,
> I'm a newbie to dipy.
>
> I was trying to follow the instructions in [1] to have dipy installed from
> the source code, so that I could execute the dipy examples.
>
> I'm using Windows 10 and Anaconda 3.
>
> When trying to execute
> python setup.py develop
>
> the Anaconda prompt yields an error that says in the end:
> File: "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126, __init__
> if self.ld_version >= "2.10.90":
> TypeError: unorderable types: NoneType() >= str()
>
> I've been googling for a solution without success.
>
> I don't know whether this looks like Anaconda3 is trying to use cygwin
> instead of mingw32, and whether that is the root cause.
>
> In either case, does anyone know how to solve the issue?
>
> Attached is the trace (it's short) of the error if this is of any help.
>
>
> Thank you,
> JON HAITZ
>
>
> [1] http://nipy.org/dipy/installation.html#install-source-nix

I'm sorry, I'm afraid I don't personally use Anaconda, so I have no
experience of fixing compilation errors on Anaconda.   Ariel - have
you come across this?

You could also try on the anaconda support channels (issues, mailing
list) - it may well be a general problem rather than one specific to
dipy,

Best,

Matthew

From arokem at gmail.com  Sat May 21 10:55:15 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Sat, 21 May 2016 07:55:15 -0700
Subject: [Neuroimaging] Fwd: [dipy] Issues trying to install dipy
In-Reply-To: <CAH6Pt5rVree2gyERp3DSXVmC_dS4j4gQ66YJpouO62W4JpvX=g@mail.gmail.com>
References: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>
 <CAAKhGh9-emozuiqQv2KpY8hghks-1b-bid1PQufJ7YUwDffoDQ@mail.gmail.com>
 <CAH6Pt5rVree2gyERp3DSXVmC_dS4j4gQ66YJpouO62W4JpvX=g@mail.gmail.com>
Message-ID: <CA+X4w0zbG_+5FQx8ALCvij54y+p_4WyfpwfD3LyigWoxAqH66A@mail.gmail.com>

Hi Jon and Matthew,


On Sat, May 21, 2016 at 7:30 AM, Matthew Brett <matthew.brett at gmail.com>
wrote:

> Hi,
>
> On Sat, May 21, 2016 at 9:18 AM, Jon Haitz Legarreta
> <jhlegarreta at vicomtech.org> wrote:
> > Hi there,
> > has anybody experienced the issue below?
> >
> > Thanks,
> > JON HAITZ
> >
> >
> >
> >
> > ---------- Forwarded message ----------
> > From: Jon Haitz Legarreta <jhlegarreta at vicomtech.org>
> > Date: 18 May 2016 at 19:10
> > Subject: [dipy] Issues trying to install dipy
> > To: neuroimaging at python.org
> >
> >
> > Hi there,
> > I'm a newbie to dipy.
> >
> > I was trying to follow the instructions in [1] to have dipy installed
> from
> > the source code, so that I could execute the dipy examples.
> >
> > I'm using Windows 10 and Anaconda 3.
> >
> > When trying to execute
> > python setup.py develop
> >
> > the Anaconda prompt yields an error that says in the end:
> > File: "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126, __init__
> > if self.ld_version >= "2.10.90":
> > TypeError: unorderable types: NoneType() >= str()
> >
> > I've been googling for a solution without success.
> >
> > I don't know whether this looks like Anaconda3 is trying to use cygwin
> > instead of mingw32, and whether that is the root cause.
> >
> > In either case, does anyone know how to solve the issue?
> >
> > Attached is the trace (it's short) of the error if this is of any help.
> >
> >
> > Thank you,
> > JON HAITZ
> >
> >
> > [1] http://nipy.org/dipy/installation.html#install-source-nix
>
> I'm sorry, I'm afraid I don't personally use Anaconda, so I have no
> experience of fixing compilation errors on Anaconda.   Ariel - have
> you come across this?
>

And I don't personally use Windows...

Might this be helpful:

http://stackoverflow.com/questions/24683305/python-cant-install-packages-typeerror-unorderable-types-nonetype-str

It seems like it could be related, though it's all Greek to me.

Cheers,

Ariel


> You could also try on the anaconda support channels (issues, mailing
> list) - it may well be a general problem rather than one specific to
> dipy,
>
> Best,
>
> Matthew
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160521/f79783ec/attachment.html>

From arokem at gmail.com  Sat May 21 10:59:07 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Sat, 21 May 2016 07:59:07 -0700
Subject: [Neuroimaging] error code
In-Reply-To: <CAMYkJft5vWhhR4gRKjss3baj6vT3ABPtnoHnSGN0+Dr662RYmA@mail.gmail.com>
References: <CAMYkJft5vWhhR4gRKjss3baj6vT3ABPtnoHnSGN0+Dr662RYmA@mail.gmail.com>
Message-ID: <CA+X4w0xSGJg5KoAa0hBAva2_C9diEBXELJEhAxk8W+xHTADq=Q@mail.gmail.com>

On Fri, May 20, 2016 at 11:36 PM, Vivek Joshi <vivekjoshi1894 at gmail.com>
wrote:

> Greetings Ariel Rokem
> The following code raised the error
> import nibabel as nib
> import numpy as np
> from dipy.denoise.noise_estimate import piesno
> from dipy.data import fetch_isbi2013_2shell, read_isbi2013_2shell
>
>
> fetch_isbi2013_2shell()
> img, gtab = read_isbi2013_2shell()
> data = img.get_data()
>
> attribute error here!
>
>
This code works fine for me. Could you also send us the full error message
that you get?


> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160521/49f47cda/attachment.html>

From arokem at gmail.com  Sat May 21 10:59:57 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Sat, 21 May 2016 07:59:57 -0700
Subject: [Neuroimaging] plotting 3d dwi images
In-Reply-To: <CAMYkJftih-dMcKNi2uh=Ao+qFYmkNBoD+iHkFRx8mtbF=4RMLw@mail.gmail.com>
References: <CAMYkJftih-dMcKNi2uh=Ao+qFYmkNBoD+iHkFRx8mtbF=4RMLw@mail.gmail.com>
Message-ID: <CA+X4w0yOr5s+ZqX0TcoZipYohOmWP+zSDGd+kYqo7JxW0wpZkw@mail.gmail.com>

On Fri, May 20, 2016 at 11:39 PM, Vivek Joshi <vivekjoshi1894 at gmail.com>
wrote:

> Is there a way to plot a histogram of an original 3d image like
> sherbrooke_3shell and its denoised image so that those two can be compared
> and denoising can be witnessed.
>
>
The following example shows something like that:

http://nipy.org/dipy/examples_built/denoise_nlmeans.html#example-denoise-nlmeans



> Thank You!
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160521/03a4be0a/attachment.html>

From jhlegarreta at vicomtech.org  Mon May 23 18:19:57 2016
From: jhlegarreta at vicomtech.org (Jon Haitz Legarreta)
Date: Tue, 24 May 2016 00:19:57 +0200
Subject: [Neuroimaging] Fwd: [dipy] Issues trying to install dipy
In-Reply-To: <CA+X4w0zbG_+5FQx8ALCvij54y+p_4WyfpwfD3LyigWoxAqH66A@mail.gmail.com>
References: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>
 <CAAKhGh9-emozuiqQv2KpY8hghks-1b-bid1PQufJ7YUwDffoDQ@mail.gmail.com>
 <CAH6Pt5rVree2gyERp3DSXVmC_dS4j4gQ66YJpouO62W4JpvX=g@mail.gmail.com>
 <CA+X4w0zbG_+5FQx8ALCvij54y+p_4WyfpwfD3LyigWoxAqH66A@mail.gmail.com>
Message-ID: <CAAKhGh_7yUqHe7EVVmpNvNSbZuv1029CfjvK8Vb3Od57FRssEg@mail.gmail.com>

Hi there,
thank you Matthew and Ariel.

The link pointed by Ariel does not seem to be a solution; after having
installed MinGW, as suggested in the link and although I'm aware it might
be unnecessary, the Anaconda3 powershell still yields a similar error, now
pointing to MSVC (which I do not have on my system):

  File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 157, in
__init__
    self.dll_libraries = get_msvcr()
  File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 86, in
get_msvcr
    raise ValueError("Unknown MS Compiler version %s " % msc_ver)
ValueError: Unknown MS Compiler version 1900

I'll try to investigate further, and will let you know.

Kind regards,
JON HAITZ





On 21 May 2016 at 16:55, Ariel Rokem <arokem at gmail.com> wrote:

> Hi Jon and Matthew,
>
>
> On Sat, May 21, 2016 at 7:30 AM, Matthew Brett <matthew.brett at gmail.com>
> wrote:
>
>> Hi,
>>
>> On Sat, May 21, 2016 at 9:18 AM, Jon Haitz Legarreta
>> <jhlegarreta at vicomtech.org> wrote:
>> > Hi there,
>> > has anybody experienced the issue below?
>> >
>> > Thanks,
>> > JON HAITZ
>> >
>> >
>> >
>> >
>> > ---------- Forwarded message ----------
>> > From: Jon Haitz Legarreta <jhlegarreta at vicomtech.org>
>> > Date: 18 May 2016 at 19:10
>> > Subject: [dipy] Issues trying to install dipy
>> > To: neuroimaging at python.org
>> >
>> >
>> > Hi there,
>> > I'm a newbie to dipy.
>> >
>> > I was trying to follow the instructions in [1] to have dipy installed
>> from
>> > the source code, so that I could execute the dipy examples.
>> >
>> > I'm using Windows 10 and Anaconda 3.
>> >
>> > When trying to execute
>> > python setup.py develop
>> >
>> > the Anaconda prompt yields an error that says in the end:
>> > File: "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126,
>> __init__
>> > if self.ld_version >= "2.10.90":
>> > TypeError: unorderable types: NoneType() >= str()
>> >
>> > I've been googling for a solution without success.
>> >
>> > I don't know whether this looks like Anaconda3 is trying to use cygwin
>> > instead of mingw32, and whether that is the root cause.
>> >
>> > In either case, does anyone know how to solve the issue?
>> >
>> > Attached is the trace (it's short) of the error if this is of any help.
>> >
>> >
>> > Thank you,
>> > JON HAITZ
>> >
>> >
>> > [1] http://nipy.org/dipy/installation.html#install-source-nix
>>
>> I'm sorry, I'm afraid I don't personally use Anaconda, so I have no
>> experience of fixing compilation errors on Anaconda.   Ariel - have
>> you come across this?
>>
>
> And I don't personally use Windows...
>
> Might this be helpful:
>
>
> http://stackoverflow.com/questions/24683305/python-cant-install-packages-typeerror-unorderable-types-nonetype-str
>
> It seems like it could be related, though it's all Greek to me.
>
> Cheers,
>
> Ariel
>
>
>> You could also try on the anaconda support channels (issues, mailing
>> list) - it may well be a general problem rather than one specific to
>> dipy,
>>
>> Best,
>>
>> Matthew
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160524/558ead98/attachment.html>

From arokem at gmail.com  Mon May 23 19:08:06 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Mon, 23 May 2016 16:08:06 -0700
Subject: [Neuroimaging] Fwd: [dipy] Issues trying to install dipy
In-Reply-To: <CAAKhGh_7yUqHe7EVVmpNvNSbZuv1029CfjvK8Vb3Od57FRssEg@mail.gmail.com>
References: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>
 <CAAKhGh9-emozuiqQv2KpY8hghks-1b-bid1PQufJ7YUwDffoDQ@mail.gmail.com>
 <CAH6Pt5rVree2gyERp3DSXVmC_dS4j4gQ66YJpouO62W4JpvX=g@mail.gmail.com>
 <CA+X4w0zbG_+5FQx8ALCvij54y+p_4WyfpwfD3LyigWoxAqH66A@mail.gmail.com>
 <CAAKhGh_7yUqHe7EVVmpNvNSbZuv1029CfjvK8Vb3Od57FRssEg@mail.gmail.com>
Message-ID: <CA+X4w0z=1qjsbLcknNSjMZOkAjrSxhE=8B23E4+_YJPjcJTVrw@mail.gmail.com>

On Mon, May 23, 2016 at 3:19 PM, Jon Haitz Legarreta <
jhlegarreta at vicomtech.org> wrote:

> Hi there,
> thank you Matthew and Ariel.
>
> The link pointed by Ariel does not seem to be a solution; after having
> installed MinGW, as suggested in the link and although I'm aware it might
> be unnecessary, the Anaconda3 powershell still yields a similar error, now
> pointing to MSVC (which I do not have on my system):
>
>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 157, in
> __init__
>     self.dll_libraries = get_msvcr()
>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 86, in
> get_msvcr
>     raise ValueError("Unknown MS Compiler version %s " % msc_ver)
> ValueError: Unknown MS Compiler version 1900
>
> Looks like maybe you ran into this corner case?

http://stackoverflow.com/a/34427014/3532933





> I'll try to investigate further, and will let you know.
>
> Kind regards,
> JON HAITZ
>
>
>
>
>
>
> On 21 May 2016 at 16:55, Ariel Rokem <arokem at gmail.com> wrote:
>
>> Hi Jon and Matthew,
>>
>>
>> On Sat, May 21, 2016 at 7:30 AM, Matthew Brett <matthew.brett at gmail.com>
>> wrote:
>>
>>> Hi,
>>>
>>> On Sat, May 21, 2016 at 9:18 AM, Jon Haitz Legarreta
>>> <jhlegarreta at vicomtech.org> wrote:
>>> > Hi there,
>>> > has anybody experienced the issue below?
>>> >
>>> > Thanks,
>>> > JON HAITZ
>>> >
>>> >
>>> >
>>> >
>>> > ---------- Forwarded message ----------
>>> > From: Jon Haitz Legarreta <jhlegarreta at vicomtech.org>
>>> > Date: 18 May 2016 at 19:10
>>> > Subject: [dipy] Issues trying to install dipy
>>> > To: neuroimaging at python.org
>>> >
>>> >
>>> > Hi there,
>>> > I'm a newbie to dipy.
>>> >
>>> > I was trying to follow the instructions in [1] to have dipy installed
>>> from
>>> > the source code, so that I could execute the dipy examples.
>>> >
>>> > I'm using Windows 10 and Anaconda 3.
>>> >
>>> > When trying to execute
>>> > python setup.py develop
>>> >
>>> > the Anaconda prompt yields an error that says in the end:
>>> > File: "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126,
>>> __init__
>>> > if self.ld_version >= "2.10.90":
>>> > TypeError: unorderable types: NoneType() >= str()
>>> >
>>> > I've been googling for a solution without success.
>>> >
>>> > I don't know whether this looks like Anaconda3 is trying to use cygwin
>>> > instead of mingw32, and whether that is the root cause.
>>> >
>>> > In either case, does anyone know how to solve the issue?
>>> >
>>> > Attached is the trace (it's short) of the error if this is of any help.
>>> >
>>> >
>>> > Thank you,
>>> > JON HAITZ
>>> >
>>> >
>>> > [1] http://nipy.org/dipy/installation.html#install-source-nix
>>>
>>> I'm sorry, I'm afraid I don't personally use Anaconda, so I have no
>>> experience of fixing compilation errors on Anaconda.   Ariel - have
>>> you come across this?
>>>
>>
>> And I don't personally use Windows...
>>
>> Might this be helpful:
>>
>>
>> http://stackoverflow.com/questions/24683305/python-cant-install-packages-typeerror-unorderable-types-nonetype-str
>>
>> It seems like it could be related, though it's all Greek to me.
>>
>> Cheers,
>>
>> Ariel
>>
>>
>>> You could also try on the anaconda support channels (issues, mailing
>>> list) - it may well be a general problem rather than one specific to
>>> dipy,
>>>
>>> Best,
>>>
>>> Matthew
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160523/1c3d1c2f/attachment.html>

From avesani at fbk.eu  Wed May 25 05:31:11 2016
From: avesani at fbk.eu (Paolo Avesani)
Date: Wed, 25 May 2016 11:31:11 +0200
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
Message-ID: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>

I would like to take advantage of the "predict" method of reconstruction
models in dipy. The goal is to assess the quality of results.

I have already computed the reconstruction models using mrtrix3 and stored
the ODF files. For this reason I would need to initialize the csd model by
importing the data from ODF stored by mrtrix3.

The questions are manifold:
- may I initialize the csd model by providing the precomputed values and
skipping the "fit" step?
- may I import the value of precomputed model from a file stored by mrtrix3?
- is the csd model in dipy compliant with the output of multi-shell csd
model computed by mrtrix3?

I hope my questions and my goal is formulated clearly.
Thanks for your support.
Paolo
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160525/3189e485/attachment.html>

From mrbago at gmail.com  Wed May 25 16:09:53 2016
From: mrbago at gmail.com (Bago)
Date: Wed, 25 May 2016 20:09:53 +0000
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
In-Reply-To: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
References: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
Message-ID: <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>

Hi Paolo,
  mrtrix and dipy define the SH basis slightly differently, so the
precomputed FOD values need to be adjusted if you want to skip the fit step
and initialize the Fit object directly. IRC we don't currently have the
code to do that, but it would be something we'd like to incorporate.

I have a WIP version of the multi-shell CSD model on a separate branch, I
plan on merging it but wasn't intending to get to that for a few months. If
you'd like to look at before then I can push the branch up to github.

Bago

On Wed, May 25, 2016 at 2:31 AM Paolo Avesani <avesani at fbk.eu> wrote:

> I would like to take advantage of the "predict" method of reconstruction
> models in dipy. The goal is to assess the quality of results.
>
> I have already computed the reconstruction models using mrtrix3 and stored
> the ODF files. For this reason I would need to initialize the csd model by
> importing the data from ODF stored by mrtrix3.
>
> The questions are manifold:
> - may I initialize the csd model by providing the precomputed values and
> skipping the "fit" step?
> - may I import the value of precomputed model from a file stored by
> mrtrix3?
> - is the csd model in dipy compliant with the output of multi-shell csd
> model computed by mrtrix3?
>
> I hope my questions and my goal is formulated clearly.
> Thanks for your support.
> Paolo
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160525/2db6736d/attachment.html>

From arokem at gmail.com  Wed May 25 18:24:48 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Wed, 25 May 2016 15:24:48 -0700
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
In-Reply-To: <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>
References: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
 <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>
Message-ID: <CA+X4w0xfBuK8yy8g=RdXQbF503jtHw1PaNOE8Bs0pH7BQq2Z9A@mail.gmail.com>

On Wed, May 25, 2016 at 1:09 PM, Bago <mrbago at gmail.com> wrote:

> Hi Paolo,
>   mrtrix and dipy define the SH basis slightly differently, so the
> precomputed FOD values need to be adjusted if you want to skip the fit step
> and initialize the Fit object directly. IRC we don't currently have the
> code to do that, but it would be something we'd like to incorporate.
>
> Did they change their basis set when they transitioned to mrtrix3? We do
have these functions:

https://github.com/nipy/dipy/blob/master/dipy/reconst/shm.py#L852-L923

That should work with the previous version of mrtrix (mrtrix2?). You can
use these to transform between coefficient sets:

    sf = sh_to_sf(mrtrix_coeffs, sphere, sh_order, basis_type='mrtrix')
    dipy_coeffs = sf_to_sh(sf, sphere, sh_order, basis_type=None) # This
defaults to the dipy basis set

and then use the CSD model object to predict:

    from dipy.reconst.csdeconv import  ConstrainedSphericalDeconvModel
    csd_model = ConstrainedSphericalDeconvModel(gtab, response,
sh_order=sh_order) # Note: you still need to calculate the response
function!
    pred_signal = csd_model.predict(dipy_coeffs, gtab, S0)

I think that something like this should work (but I haven't tried it
myself).


> I have a WIP version of the multi-shell CSD model on a separate branch, I
> plan on merging it but wasn't intending to get to that for a few months. If
> you'd like to look at before then I can push the branch up to github.
>
> Sounds interesting! I'd love to see what you have so far!

Cheers,

Ariel


> Bago
>
> On Wed, May 25, 2016 at 2:31 AM Paolo Avesani <avesani at fbk.eu> wrote:
>
>> I would like to take advantage of the "predict" method of reconstruction
>> models in dipy. The goal is to assess the quality of results.
>>
>> I have already computed the reconstruction models using mrtrix3 and
>> stored the ODF files. For this reason I would need to initialize the csd
>> model by importing the data from ODF stored by mrtrix3.
>>
>> The questions are manifold:
>> - may I initialize the csd model by providing the precomputed values and
>> skipping the "fit" step?
>> - may I import the value of precomputed model from a file stored by
>> mrtrix3?
>> - is the csd model in dipy compliant with the output of multi-shell csd
>> model computed by mrtrix3?
>>
>> I hope my questions and my goal is formulated clearly.
>> Thanks for your support.
>> Paolo
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160525/59812369/attachment.html>

From mrbago at gmail.com  Wed May 25 19:41:41 2016
From: mrbago at gmail.com (Bago)
Date: Wed, 25 May 2016 23:41:41 +0000
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
In-Reply-To: <CA+X4w0xfBuK8yy8g=RdXQbF503jtHw1PaNOE8Bs0pH7BQq2Z9A@mail.gmail.com>
References: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
 <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>
 <CA+X4w0xfBuK8yy8g=RdXQbF503jtHw1PaNOE8Bs0pH7BQq2Z9A@mail.gmail.com>
Message-ID: <CAHHh6CbGtSCz7HCarvR6_qHBMX-zrNwKfuY9k6N=svc4BHux_A@mail.gmail.com>

I believe they did change their basis (please correct me if I'm wrong but I
believe they went from a non-normalized SH basis to a normalized SH basis).

Also projecting onto a sphere is one way to _estimate_ the coefficients in
a different basis. The cleaner way is to just re-order the coefficients and
apply the appropriate scaling. If both basis are normalized (which dipy is)
the scaling should be 1 or -1.

Bago

On Wed, May 25, 2016 at 3:31 PM Ariel Rokem <arokem at gmail.com> wrote:

> On Wed, May 25, 2016 at 1:09 PM, Bago <mrbago at gmail.com> wrote:
>
>> Hi Paolo,
>>   mrtrix and dipy define the SH basis slightly differently, so the
>> precomputed FOD values need to be adjusted if you want to skip the fit step
>> and initialize the Fit object directly. IRC we don't currently have the
>> code to do that, but it would be something we'd like to incorporate.
>>
>> Did they change their basis set when they transitioned to mrtrix3? We do
> have these functions:
>
> https://github.com/nipy/dipy/blob/master/dipy/reconst/shm.py#L852-L923
>
> That should work with the previous version of mrtrix (mrtrix2?). You can
> use these to transform between coefficient sets:
>
>     sf = sh_to_sf(mrtrix_coeffs, sphere, sh_order, basis_type='mrtrix')
>     dipy_coeffs = sf_to_sh(sf, sphere, sh_order, basis_type=None) # This
> defaults to the dipy basis set
>
> and then use the CSD model object to predict:
>
>     from dipy.reconst.csdeconv import  ConstrainedSphericalDeconvModel
>     csd_model = ConstrainedSphericalDeconvModel(gtab, response,
> sh_order=sh_order) # Note: you still need to calculate the response
> function!
>     pred_signal = csd_model.predict(dipy_coeffs, gtab, S0)
>
> I think that something like this should work (but I haven't tried it
> myself).
>
>
>> I have a WIP version of the multi-shell CSD model on a separate branch, I
>> plan on merging it but wasn't intending to get to that for a few months. If
>> you'd like to look at before then I can push the branch up to github.
>>
>> Sounds interesting! I'd love to see what you have so far!
>
> Cheers,
>
> Ariel
>
>
>> Bago
>>
>> On Wed, May 25, 2016 at 2:31 AM Paolo Avesani <avesani at fbk.eu> wrote:
>>
>>> I would like to take advantage of the "predict" method of reconstruction
>>> models in dipy. The goal is to assess the quality of results.
>>>
>>> I have already computed the reconstruction models using mrtrix3 and
>>> stored the ODF files. For this reason I would need to initialize the csd
>>> model by importing the data from ODF stored by mrtrix3.
>>>
>>> The questions are manifold:
>>> - may I initialize the csd model by providing the precomputed values and
>>> skipping the "fit" step?
>>> - may I import the value of precomputed model from a file stored by
>>> mrtrix3?
>>> - is the csd model in dipy compliant with the output of multi-shell csd
>>> model computed by mrtrix3?
>>>
>>> I hope my questions and my goal is formulated clearly.
>>> Thanks for your support.
>>> Paolo
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160525/7d60c312/attachment.html>

From arokem at gmail.com  Wed May 25 19:55:02 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Wed, 25 May 2016 16:55:02 -0700
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
In-Reply-To: <CAHHh6CbGtSCz7HCarvR6_qHBMX-zrNwKfuY9k6N=svc4BHux_A@mail.gmail.com>
References: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
 <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>
 <CA+X4w0xfBuK8yy8g=RdXQbF503jtHw1PaNOE8Bs0pH7BQq2Z9A@mail.gmail.com>
 <CAHHh6CbGtSCz7HCarvR6_qHBMX-zrNwKfuY9k6N=svc4BHux_A@mail.gmail.com>
Message-ID: <CA+X4w0xwbHP9O-eH1WvyaU8ncLb4arZF_j8FvEkKZexXoH3FjA@mail.gmail.com>

On Wed, May 25, 2016 at 4:41 PM, Bago <mrbago at gmail.com> wrote:

> I believe they did change their basis (please correct me if I'm wrong but
> I believe they went from a non-normalized SH basis to a normalized SH
> basis).
>
>
So they have the same basis as dipy now, but the coefficients appear in a
different order? That should make life even easier!


> Also projecting onto a sphere is one way to _estimate_ the coefficients in
> a different basis. The cleaner way is to just re-order the coefficients and
> apply the appropriate scaling. If both basis are normalized (which dipy is)
> the scaling should be 1 or -1.
>
>
Fair point, but to be just a little bit facetious: given enough points on
the sphere and knowledge of the target maximal order of the coefficients,
wouldn't estimating be the same as transforming? Works for the FFT, I
believe :-)

Bago
>
> On Wed, May 25, 2016 at 3:31 PM Ariel Rokem <arokem at gmail.com> wrote:
>
>> On Wed, May 25, 2016 at 1:09 PM, Bago <mrbago at gmail.com> wrote:
>>
>>> Hi Paolo,
>>>   mrtrix and dipy define the SH basis slightly differently, so the
>>> precomputed FOD values need to be adjusted if you want to skip the fit step
>>> and initialize the Fit object directly. IRC we don't currently have the
>>> code to do that, but it would be something we'd like to incorporate.
>>>
>>> Did they change their basis set when they transitioned to mrtrix3? We do
>> have these functions:
>>
>> https://github.com/nipy/dipy/blob/master/dipy/reconst/shm.py#L852-L923
>>
>> That should work with the previous version of mrtrix (mrtrix2?). You can
>> use these to transform between coefficient sets:
>>
>>     sf = sh_to_sf(mrtrix_coeffs, sphere, sh_order, basis_type='mrtrix')
>>     dipy_coeffs = sf_to_sh(sf, sphere, sh_order, basis_type=None) # This
>> defaults to the dipy basis set
>>
>> and then use the CSD model object to predict:
>>
>>     from dipy.reconst.csdeconv import  ConstrainedSphericalDeconvModel
>>     csd_model = ConstrainedSphericalDeconvModel(gtab, response,
>> sh_order=sh_order) # Note: you still need to calculate the response
>> function!
>>     pred_signal = csd_model.predict(dipy_coeffs, gtab, S0)
>>
>> I think that something like this should work (but I haven't tried it
>> myself).
>>
>>
>>> I have a WIP version of the multi-shell CSD model on a separate branch,
>>> I plan on merging it but wasn't intending to get to that for a few months.
>>> If you'd like to look at before then I can push the branch up to github.
>>>
>>> Sounds interesting! I'd love to see what you have so far!
>>
>> Cheers,
>>
>> Ariel
>>
>>
>>> Bago
>>>
>>> On Wed, May 25, 2016 at 2:31 AM Paolo Avesani <avesani at fbk.eu> wrote:
>>>
>>>> I would like to take advantage of the "predict" method of
>>>> reconstruction models in dipy. The goal is to assess the quality of results.
>>>>
>>>> I have already computed the reconstruction models using mrtrix3 and
>>>> stored the ODF files. For this reason I would need to initialize the csd
>>>> model by importing the data from ODF stored by mrtrix3.
>>>>
>>>> The questions are manifold:
>>>> - may I initialize the csd model by providing the precomputed values
>>>> and skipping the "fit" step?
>>>> - may I import the value of precomputed model from a file stored by
>>>> mrtrix3?
>>>> - is the csd model in dipy compliant with the output of multi-shell csd
>>>> model computed by mrtrix3?
>>>>
>>>> I hope my questions and my goal is formulated clearly.
>>>> Thanks for your support.
>>>> Paolo
>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160525/8cc87383/attachment.html>

From stjeansam at gmail.com  Wed May 25 20:12:57 2016
From: stjeansam at gmail.com (Samuel St-Jean)
Date: Thu, 26 May 2016 02:12:57 +0200
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
In-Reply-To: <CA+X4w0xwbHP9O-eH1WvyaU8ncLb4arZF_j8FvEkKZexXoH3FjA@mail.gmail.com>
References: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
 <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>
 <CA+X4w0xfBuK8yy8g=RdXQbF503jtHw1PaNOE8Bs0pH7BQq2Z9A@mail.gmail.com>
 <CAHHh6CbGtSCz7HCarvR6_qHBMX-zrNwKfuY9k6N=svc4BHux_A@mail.gmail.com>
 <CA+X4w0xwbHP9O-eH1WvyaU8ncLb4arZF_j8FvEkKZexXoH3FjA@mail.gmail.com>
Message-ID: <CAKADGuq6kJRd9q1d9zv7=ggXGAau=3-HxoWXz73_qgxuEUTeKQ@mail.gmail.com>

Their wiki explains it, a sqrt(2) to normalize is used in mrtrix3, so
multiplying your coefficients with that and using the mrtrix2 functions
should do it.

Although dipy only does single shell, so conclude with consideration that
the algorithm is different from mrtrix3. Also, csd is a bad signal
predictor (but good for angle estimation), see the sparc dmri challenge
paper for example.
On May 26, 2016 07:55, "Ariel Rokem" <arokem at gmail.com> wrote:

>
>
> On Wed, May 25, 2016 at 4:41 PM, Bago <mrbago at gmail.com> wrote:
>
>> I believe they did change their basis (please correct me if I'm wrong but
>> I believe they went from a non-normalized SH basis to a normalized SH
>> basis).
>>
>>
> So they have the same basis as dipy now, but the coefficients appear in a
> different order? That should make life even easier!
>
>
>> Also projecting onto a sphere is one way to _estimate_ the coefficients
>> in a different basis. The cleaner way is to just re-order the coefficients
>> and apply the appropriate scaling. If both basis are normalized (which dipy
>> is) the scaling should be 1 or -1.
>>
>>
> Fair point, but to be just a little bit facetious: given enough points on
> the sphere and knowledge of the target maximal order of the coefficients,
> wouldn't estimating be the same as transforming? Works for the FFT, I
> believe :-)
>
> Bago
>>
>> On Wed, May 25, 2016 at 3:31 PM Ariel Rokem <arokem at gmail.com> wrote:
>>
>>> On Wed, May 25, 2016 at 1:09 PM, Bago <mrbago at gmail.com> wrote:
>>>
>>>> Hi Paolo,
>>>>   mrtrix and dipy define the SH basis slightly differently, so the
>>>> precomputed FOD values need to be adjusted if you want to skip the fit step
>>>> and initialize the Fit object directly. IRC we don't currently have the
>>>> code to do that, but it would be something we'd like to incorporate.
>>>>
>>>> Did they change their basis set when they transitioned to mrtrix3? We
>>> do have these functions:
>>>
>>> https://github.com/nipy/dipy/blob/master/dipy/reconst/shm.py#L852-L923
>>>
>>> That should work with the previous version of mrtrix (mrtrix2?). You can
>>> use these to transform between coefficient sets:
>>>
>>>     sf = sh_to_sf(mrtrix_coeffs, sphere, sh_order, basis_type='mrtrix')
>>>     dipy_coeffs = sf_to_sh(sf, sphere, sh_order, basis_type=None) # This
>>> defaults to the dipy basis set
>>>
>>> and then use the CSD model object to predict:
>>>
>>>     from dipy.reconst.csdeconv import  ConstrainedSphericalDeconvModel
>>>     csd_model = ConstrainedSphericalDeconvModel(gtab, response,
>>> sh_order=sh_order) # Note: you still need to calculate the response
>>> function!
>>>     pred_signal = csd_model.predict(dipy_coeffs, gtab, S0)
>>>
>>> I think that something like this should work (but I haven't tried it
>>> myself).
>>>
>>>
>>>> I have a WIP version of the multi-shell CSD model on a separate branch,
>>>> I plan on merging it but wasn't intending to get to that for a few months.
>>>> If you'd like to look at before then I can push the branch up to github.
>>>>
>>>> Sounds interesting! I'd love to see what you have so far!
>>>
>>> Cheers,
>>>
>>> Ariel
>>>
>>>
>>>> Bago
>>>>
>>>> On Wed, May 25, 2016 at 2:31 AM Paolo Avesani <avesani at fbk.eu> wrote:
>>>>
>>>>> I would like to take advantage of the "predict" method of
>>>>> reconstruction models in dipy. The goal is to assess the quality of results.
>>>>>
>>>>> I have already computed the reconstruction models using mrtrix3 and
>>>>> stored the ODF files. For this reason I would need to initialize the csd
>>>>> model by importing the data from ODF stored by mrtrix3.
>>>>>
>>>>> The questions are manifold:
>>>>> - may I initialize the csd model by providing the precomputed values
>>>>> and skipping the "fit" step?
>>>>> - may I import the value of precomputed model from a file stored by
>>>>> mrtrix3?
>>>>> - is the csd model in dipy compliant with the output of multi-shell
>>>>> csd model computed by mrtrix3?
>>>>>
>>>>> I hope my questions and my goal is formulated clearly.
>>>>> Thanks for your support.
>>>>> Paolo
>>>>>
>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160526/414716d2/attachment.html>

From mrbago at gmail.com  Wed May 25 21:36:51 2016
From: mrbago at gmail.com (Bago)
Date: Thu, 26 May 2016 01:36:51 +0000
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
In-Reply-To: <CAKADGuq6kJRd9q1d9zv7=ggXGAau=3-HxoWXz73_qgxuEUTeKQ@mail.gmail.com>
References: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
 <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>
 <CA+X4w0xfBuK8yy8g=RdXQbF503jtHw1PaNOE8Bs0pH7BQq2Z9A@mail.gmail.com>
 <CAHHh6CbGtSCz7HCarvR6_qHBMX-zrNwKfuY9k6N=svc4BHux_A@mail.gmail.com>
 <CA+X4w0xwbHP9O-eH1WvyaU8ncLb4arZF_j8FvEkKZexXoH3FjA@mail.gmail.com>
 <CAKADGuq6kJRd9q1d9zv7=ggXGAau=3-HxoWXz73_qgxuEUTeKQ@mail.gmail.com>
Message-ID: <CAHHh6CaU-K5tJeNXdUqdOmy09xYZ8eG36oWPFh1t8jaw7bzVVA@mail.gmail.com>

Samuel mrtrix has at least two models implemented. The multi-shell
(multi-tissue) model cannot be used with single shell data and the original
CSD model cannot be used with multi-shell data.

Bago

On Wed, May 25, 2016 at 5:13 PM Samuel St-Jean <stjeansam at gmail.com> wrote:

> Their wiki explains it, a sqrt(2) to normalize is used in mrtrix3, so
> multiplying your coefficients with that and using the mrtrix2 functions
> should do it.
>
> Although dipy only does single shell, so conclude with consideration that
> the algorithm is different from mrtrix3. Also, csd is a bad signal
> predictor (but good for angle estimation), see the sparc dmri challenge
> paper for example.
> On May 26, 2016 07:55, "Ariel Rokem" <arokem at gmail.com> wrote:
>
>>
>>
>> On Wed, May 25, 2016 at 4:41 PM, Bago <mrbago at gmail.com> wrote:
>>
>>> I believe they did change their basis (please correct me if I'm wrong
>>> but I believe they went from a non-normalized SH basis to a normalized SH
>>> basis).
>>>
>>>
>> So they have the same basis as dipy now, but the coefficients appear in a
>> different order? That should make life even easier!
>>
>>
>>> Also projecting onto a sphere is one way to _estimate_ the coefficients
>>> in a different basis. The cleaner way is to just re-order the coefficients
>>> and apply the appropriate scaling. If both basis are normalized (which dipy
>>> is) the scaling should be 1 or -1.
>>>
>>>
>> Fair point, but to be just a little bit facetious: given enough points on
>> the sphere and knowledge of the target maximal order of the coefficients,
>> wouldn't estimating be the same as transforming? Works for the FFT, I
>> believe :-)
>>
>> Bago
>>>
>>> On Wed, May 25, 2016 at 3:31 PM Ariel Rokem <arokem at gmail.com> wrote:
>>>
>>>> On Wed, May 25, 2016 at 1:09 PM, Bago <mrbago at gmail.com> wrote:
>>>>
>>>>> Hi Paolo,
>>>>>   mrtrix and dipy define the SH basis slightly differently, so the
>>>>> precomputed FOD values need to be adjusted if you want to skip the fit step
>>>>> and initialize the Fit object directly. IRC we don't currently have the
>>>>> code to do that, but it would be something we'd like to incorporate.
>>>>>
>>>>> Did they change their basis set when they transitioned to mrtrix3? We
>>>> do have these functions:
>>>>
>>>> https://github.com/nipy/dipy/blob/master/dipy/reconst/shm.py#L852-L923
>>>>
>>>> That should work with the previous version of mrtrix (mrtrix2?). You
>>>> can use these to transform between coefficient sets:
>>>>
>>>>     sf = sh_to_sf(mrtrix_coeffs, sphere, sh_order, basis_type='mrtrix')
>>>>     dipy_coeffs = sf_to_sh(sf, sphere, sh_order, basis_type=None) #
>>>> This defaults to the dipy basis set
>>>>
>>>> and then use the CSD model object to predict:
>>>>
>>>>     from dipy.reconst.csdeconv import  ConstrainedSphericalDeconvModel
>>>>     csd_model = ConstrainedSphericalDeconvModel(gtab, response,
>>>> sh_order=sh_order) # Note: you still need to calculate the response
>>>> function!
>>>>     pred_signal = csd_model.predict(dipy_coeffs, gtab, S0)
>>>>
>>>> I think that something like this should work (but I haven't tried it
>>>> myself).
>>>>
>>>>
>>>>> I have a WIP version of the multi-shell CSD model on a separate
>>>>> branch, I plan on merging it but wasn't intending to get to that for a few
>>>>> months. If you'd like to look at before then I can push the branch up to
>>>>> github.
>>>>>
>>>>> Sounds interesting! I'd love to see what you have so far!
>>>>
>>>> Cheers,
>>>>
>>>> Ariel
>>>>
>>>>
>>>>> Bago
>>>>>
>>>>> On Wed, May 25, 2016 at 2:31 AM Paolo Avesani <avesani at fbk.eu> wrote:
>>>>>
>>>>>> I would like to take advantage of the "predict" method of
>>>>>> reconstruction models in dipy. The goal is to assess the quality of results.
>>>>>>
>>>>>> I have already computed the reconstruction models using mrtrix3 and
>>>>>> stored the ODF files. For this reason I would need to initialize the csd
>>>>>> model by importing the data from ODF stored by mrtrix3.
>>>>>>
>>>>>> The questions are manifold:
>>>>>> - may I initialize the csd model by providing the precomputed values
>>>>>> and skipping the "fit" step?
>>>>>> - may I import the value of precomputed model from a file stored by
>>>>>> mrtrix3?
>>>>>> - is the csd model in dipy compliant with the output of multi-shell
>>>>>> csd model computed by mrtrix3?
>>>>>>
>>>>>> I hope my questions and my goal is formulated clearly.
>>>>>> Thanks for your support.
>>>>>> Paolo
>>>>>>
>>>>>> _______________________________________________
>>>>>> Neuroimaging mailing list
>>>>>> Neuroimaging at python.org
>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160526/21ad8df9/attachment.html>

From stjeansam at gmail.com  Wed May 25 22:07:29 2016
From: stjeansam at gmail.com (Samuel St-Jean)
Date: Thu, 26 May 2016 04:07:29 +0200
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
In-Reply-To: <CAHHh6CaU-K5tJeNXdUqdOmy09xYZ8eG36oWPFh1t8jaw7bzVVA@mail.gmail.com>
References: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
 <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>
 <CA+X4w0xfBuK8yy8g=RdXQbF503jtHw1PaNOE8Bs0pH7BQq2Z9A@mail.gmail.com>
 <CAHHh6CbGtSCz7HCarvR6_qHBMX-zrNwKfuY9k6N=svc4BHux_A@mail.gmail.com>
 <CA+X4w0xwbHP9O-eH1WvyaU8ncLb4arZF_j8FvEkKZexXoH3FjA@mail.gmail.com>
 <CAKADGuq6kJRd9q1d9zv7=ggXGAau=3-HxoWXz73_qgxuEUTeKQ@mail.gmail.com>
 <CAHHh6CaU-K5tJeNXdUqdOmy09xYZ8eG36oWPFh1t8jaw7bzVVA@mail.gmail.com>
Message-ID: <CAKADGuq-RxaTq03hNxtTr-uuib-OhKaKOi5tL_2VmQCd2_+oSg@mail.gmail.com>

Indeed, I thought I read the multitissue version was used here, my mistake
as the question does not directly imply that.

The regular single shell hopefully does the same (at least I can personally
attest the mrtrix2 version and dipy version give similar fodf up to a small
rounding factor, but that was before the cholesky decomposition step, so
now they should behave the same).
On May 26, 2016 09:37, "Bago" <mrbago at gmail.com> wrote:

> Samuel mrtrix has at least two models implemented. The multi-shell
> (multi-tissue) model cannot be used with single shell data and the original
> CSD model cannot be used with multi-shell data.
>
> Bago
>
> On Wed, May 25, 2016 at 5:13 PM Samuel St-Jean <stjeansam at gmail.com>
> wrote:
>
>> Their wiki explains it, a sqrt(2) to normalize is used in mrtrix3, so
>> multiplying your coefficients with that and using the mrtrix2 functions
>> should do it.
>>
>> Although dipy only does single shell, so conclude with consideration that
>> the algorithm is different from mrtrix3. Also, csd is a bad signal
>> predictor (but good for angle estimation), see the sparc dmri challenge
>> paper for example.
>> On May 26, 2016 07:55, "Ariel Rokem" <arokem at gmail.com> wrote:
>>
>>>
>>>
>>> On Wed, May 25, 2016 at 4:41 PM, Bago <mrbago at gmail.com> wrote:
>>>
>>>> I believe they did change their basis (please correct me if I'm wrong
>>>> but I believe they went from a non-normalized SH basis to a normalized SH
>>>> basis).
>>>>
>>>>
>>> So they have the same basis as dipy now, but the coefficients appear in
>>> a different order? That should make life even easier!
>>>
>>>
>>>> Also projecting onto a sphere is one way to _estimate_ the coefficients
>>>> in a different basis. The cleaner way is to just re-order the coefficients
>>>> and apply the appropriate scaling. If both basis are normalized (which dipy
>>>> is) the scaling should be 1 or -1.
>>>>
>>>>
>>> Fair point, but to be just a little bit facetious: given enough points
>>> on the sphere and knowledge of the target maximal order of the
>>> coefficients, wouldn't estimating be the same as transforming? Works for
>>> the FFT, I believe :-)
>>>
>>> Bago
>>>>
>>>> On Wed, May 25, 2016 at 3:31 PM Ariel Rokem <arokem at gmail.com> wrote:
>>>>
>>>>> On Wed, May 25, 2016 at 1:09 PM, Bago <mrbago at gmail.com> wrote:
>>>>>
>>>>>> Hi Paolo,
>>>>>>   mrtrix and dipy define the SH basis slightly differently, so the
>>>>>> precomputed FOD values need to be adjusted if you want to skip the fit step
>>>>>> and initialize the Fit object directly. IRC we don't currently have the
>>>>>> code to do that, but it would be something we'd like to incorporate.
>>>>>>
>>>>>> Did they change their basis set when they transitioned to mrtrix3? We
>>>>> do have these functions:
>>>>>
>>>>> https://github.com/nipy/dipy/blob/master/dipy/reconst/shm.py#L852-L923
>>>>>
>>>>> That should work with the previous version of mrtrix (mrtrix2?). You
>>>>> can use these to transform between coefficient sets:
>>>>>
>>>>>     sf = sh_to_sf(mrtrix_coeffs, sphere, sh_order, basis_type='mrtrix')
>>>>>     dipy_coeffs = sf_to_sh(sf, sphere, sh_order, basis_type=None) #
>>>>> This defaults to the dipy basis set
>>>>>
>>>>> and then use the CSD model object to predict:
>>>>>
>>>>>     from dipy.reconst.csdeconv import  ConstrainedSphericalDeconvModel
>>>>>     csd_model = ConstrainedSphericalDeconvModel(gtab, response,
>>>>> sh_order=sh_order) # Note: you still need to calculate the response
>>>>> function!
>>>>>     pred_signal = csd_model.predict(dipy_coeffs, gtab, S0)
>>>>>
>>>>> I think that something like this should work (but I haven't tried it
>>>>> myself).
>>>>>
>>>>>
>>>>>> I have a WIP version of the multi-shell CSD model on a separate
>>>>>> branch, I plan on merging it but wasn't intending to get to that for a few
>>>>>> months. If you'd like to look at before then I can push the branch up to
>>>>>> github.
>>>>>>
>>>>>> Sounds interesting! I'd love to see what you have so far!
>>>>>
>>>>> Cheers,
>>>>>
>>>>> Ariel
>>>>>
>>>>>
>>>>>> Bago
>>>>>>
>>>>>> On Wed, May 25, 2016 at 2:31 AM Paolo Avesani <avesani at fbk.eu> wrote:
>>>>>>
>>>>>>> I would like to take advantage of the "predict" method of
>>>>>>> reconstruction models in dipy. The goal is to assess the quality of results.
>>>>>>>
>>>>>>> I have already computed the reconstruction models using mrtrix3 and
>>>>>>> stored the ODF files. For this reason I would need to initialize the csd
>>>>>>> model by importing the data from ODF stored by mrtrix3.
>>>>>>>
>>>>>>> The questions are manifold:
>>>>>>> - may I initialize the csd model by providing the precomputed values
>>>>>>> and skipping the "fit" step?
>>>>>>> - may I import the value of precomputed model from a file stored by
>>>>>>> mrtrix3?
>>>>>>> - is the csd model in dipy compliant with the output of multi-shell
>>>>>>> csd model computed by mrtrix3?
>>>>>>>
>>>>>>> I hope my questions and my goal is formulated clearly.
>>>>>>> Thanks for your support.
>>>>>>> Paolo
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Neuroimaging mailing list
>>>>>>> Neuroimaging at python.org
>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Neuroimaging mailing list
>>>>>> Neuroimaging at python.org
>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>
>>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>>
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160526/35ae3eea/attachment-0001.html>

From marc.cote.19 at gmail.com  Thu May 26 08:07:32 2016
From: marc.cote.19 at gmail.com (Marc)
Date: Thu, 26 May 2016 08:07:32 -0400
Subject: [Neuroimaging] NiBabel - Streamlines API - TRK header extension
Message-ID: <5746E704.5040100@gmail.com>

Hi everyone,

I'm currently working on an API for streamlines in NiBabel and I would 
like the feedback of the Nipy community about an unofficial extension 
for the TRK header.

Right now, the TrackVis's file format for streamlines supports 
conserving additional information for each point and/or streamline. The 
format refers to that additional information as "scalars" and 
"properties" respectively (ref: 
http://www.trackvis.org/docs/?subsect=fileformat). There is a field in 
the header to keep a tag name for the first 20 scalars and the first 20 
properties even though the format supports having up to 2^16 scalars and 
2^16 properties.

First, 20 is not a lot when you think of all the metrics one could want 
to keep (e.g. color, FA, curvature, torsion, MD, cluster ID, etc.) in 
its tractogram file. On top of that, some of these metrics consist of 
more than one value, for instance the color is in fact composed of three 
values (i.e. RGB). I find it a bit wasteful to use up multiple tag names 
only because it is composed of multiple values.

What I proposed is to encode that number of values in the tag name at 
save time and take that encoded number into account when loading a TRK 
file. I have two ways in mind and I would like your opinion.

*1st approach*
A tag name can only have 20 characters/bytes. I would use the last two 
bytes to encode the number of values associated to a given tag name. The 
first byte would always be \x00 (EOL) so that TrackVis doesn't try to 
interpret the last byte which will be set to the number of values (since 
it is a uint8 the range is [0, 255]). The upside of this approach is 
that TrackVis is not aware (read doesn't crash) thanks to the \x00 
"hack". The downside is that a TrackVis's user will be less aware of 
what's going on behind the scene but a NiBabel's user won't see a 
difference since we revert the process when loading the TRK file.

*2nd approach
*This consists of simply adding the number of values in plain ascii text 
as follows: "property_name-n=3" (or a better naming convention if you 
have one). The upside is the TrackVis's user will see how many values a 
given scalar/property is composed of. The downside is it will use up 
more valuable characters as the number of values gets bigger.

That said, no matter the approach, there is another small downside. 
Coming back to the color example, TrackVis will associate the tag name 
only for the first value (R) and the remaining two values (GB) will have 
an empty tag name. The following picture shows a TRK file, written using 
the first approach, that has been loaded in TrackVis. The properties 
(values per streamline) saved are a color (RGB), the mean curvature and 
the mean torsion of a streamline.




For more information see the following PR: 
https://github.com/nipy/nibabel/pull/391/

Thanks for your feedback.

-- 
Marc

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160526/40c76ca5/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 24501 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160526/40c76ca5/attachment.png>

From jean.christophe.houde at gmail.com  Thu May 26 09:56:19 2016
From: jean.christophe.houde at gmail.com (Jean-Christophe Houde)
Date: Thu, 26 May 2016 09:56:19 -0400
Subject: [Neuroimaging] NiBabel - Streamlines API - TRK header extension
In-Reply-To: <5746E704.5040100@gmail.com>
References: <5746E704.5040100@gmail.com>
Message-ID: <CADoUENNvNWUs5-ZvyNbivQFNo_tn295cvFsd6fsqM7eomJQEhQ@mail.gmail.com>

Hi Marc-Alex,

a few small things:

- you say you can save 20 property tags and 20  scalar tags. However, in
the trk format description you linked, it seems to say only 10 for each.

- For your example file, I guess that means you'd be saving using the hack
so you know that the first 3 properties written for a track are the RGB
components, and the follwoing 2 are the mean curvature and mean torsion.
Just to make sure I understand, that means that you would set n_properties
to 5, to reflect the total number?

If you can share your example file, I'd like to load it in our soft to
check the compatibility (pretty sure we would be willing to support that).

--
JC

2016-05-26 8:07 GMT-04:00 Marc <marc.cote.19 at gmail.com>:

> Hi everyone,
>
> I'm currently working on an API for streamlines in NiBabel and I would
> like the feedback of the Nipy community about an unofficial extension for
> the TRK header.
>
> Right now, the TrackVis's file format for streamlines supports conserving
> additional information for each point and/or streamline. The format refers
> to that additional information as "scalars" and "properties" respectively
> (ref: http://www.trackvis.org/docs/?subsect=fileformat). There is a field
> in the header to keep a tag name for the first 20 scalars and the first 20
> properties even though the format supports having up to 2^16 scalars and
> 2^16 properties.
>
> First, 20 is not a lot when you think of all the metrics one could want to
> keep (e.g. color, FA, curvature, torsion, MD, cluster ID, etc.) in its
> tractogram file. On top of that, some of these metrics consist of more than
> one value, for instance the color is in fact composed of three values (i.e.
> RGB). I find it a bit wasteful to use up multiple tag names only because it
> is composed of multiple values.
>
> What I proposed is to encode that number of values in the tag name at save
> time and take that encoded number into account when loading a TRK file. I
> have two ways in mind and I would like your opinion.
>
> *1st approach*
> A tag name can only have 20 characters/bytes. I would use the last two
> bytes to encode the number of values associated to a given tag name. The
> first byte would always be \x00 (EOL) so that TrackVis doesn't try to
> interpret the last byte which will be set to the number of values (since it
> is a uint8 the range is [0, 255]). The upside of this approach is that
> TrackVis is not aware (read doesn't crash) thanks to the \x00 "hack". The
> downside is that a TrackVis's user will be less aware of what's going on
> behind the scene but a NiBabel's user won't see a difference since we
> revert the process when loading the TRK file.
>
>
> *2nd approach *This consists of simply adding the number of values in
> plain ascii text as follows: "property_name-n=3" (or a better naming
> convention if you have one). The upside is the TrackVis's user will see how
> many values a given scalar/property is composed of. The downside is it will
> use up more valuable characters as the number of values gets bigger.
>
> That said, no matter the approach, there is another small downside. Coming
> back to the color example, TrackVis will associate the tag name only for
> the first value (R) and the remaining two values (GB) will have an empty
> tag name. The following picture shows a TRK file, written using the first
> approach, that has been loaded in TrackVis. The properties (values per
> streamline) saved are a color (RGB), the mean curvature and the mean
> torsion of a streamline.
>
>
>
>
> For more information see the following PR:
> <https://github.com/nipy/nibabel/pull/391/>
> https://github.com/nipy/nibabel/pull/391/
>
> Thanks for your feedback.
>
> --
> Marc
>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160526/09f3dafd/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 24501 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160526/09f3dafd/attachment.png>

From marc.cote.19 at gmail.com  Thu May 26 11:15:41 2016
From: marc.cote.19 at gmail.com (=?UTF-8?B?TWFyYy1BbGV4YW5kcmUgQ8O0dMOp?=)
Date: Thu, 26 May 2016 11:15:41 -0400
Subject: [Neuroimaging] NiBabel - Streamlines API - TRK header extension
In-Reply-To: <CADoUENNvNWUs5-ZvyNbivQFNo_tn295cvFsd6fsqM7eomJQEhQ@mail.gmail.com>
References: <5746E704.5040100@gmail.com>
 <CADoUENNvNWUs5-ZvyNbivQFNo_tn295cvFsd6fsqM7eomJQEhQ@mail.gmail.com>
Message-ID: <5747131D.5070602@gmail.com>

On 16-05-26 09:56 AM, Jean-Christophe Houde wrote:
> Hi Marc-Alex,
>
> a few small things:
>
> - you say you can save 20 property tags and 20  scalar tags. However, 
> in the trk format description you linked, it seems to say only 10 for 
> each.
You are right, it is 10 for each not 20.
>
> - For your example file, I guess that means you'd be saving using the 
> hack so you know that the first 3 properties written for a track are 
> the RGB components, and the follwoing 2 are the mean curvature and 
> mean torsion. Just to make sure I understand, that means that you 
> would set n_properties to 5, to reflect the total number?
Yes, the n_properties would be set to the total number of values 
associated to properties, in this case 5.
>
> If you can share your example file, I'd like to load it in our soft to 
> check the compatibility (pretty sure we would be willing to support that).
The file I used is the following:
https://github.com/MarcCote/nibabel/blob/streamlines_api/nibabel/streamlines/tests/data/complex.trk
Note that this file contains 3 streamlines composed of 1, 2 and 3 random 
points respectively, so they might not be displayable.

Also, regarding your software, can you check that
https://github.com/MarcCote/nibabel/blob/streamlines_api/nibabel/streamlines/tests/data/standard.trk
is displayed correctly on top of
https://github.com/MarcCote/nibabel/blob/streamlines_api/nibabel/streamlines/tests/data/standard.nii.gz
as expected (I tested it with TrackVis and dipy.viz and both are 
displaying it correctly).
>
> --
> JC
>
> 2016-05-26 8:07 GMT-04:00 Marc <marc.cote.19 at gmail.com 
> <mailto:marc.cote.19 at gmail.com>>:
>
>     Hi everyone,
>
>     I'm currently working on an API for streamlines in NiBabel and I
>     would like the feedback of the Nipy community about an unofficial
>     extension for the TRK header.
>
>     Right now, the TrackVis's file format for streamlines supports
>     conserving additional information for each point and/or
>     streamline. The format refers to that additional information as
>     "scalars" and "properties" respectively (ref:
>     http://www.trackvis.org/docs/?subsect=fileformat). There is a
>     field in the header to keep a tag name for the first 20 scalars
>     and the first 20 properties even though the format supports having
>     up to 2^16 scalars and 2^16 properties.
>
>     First, 20 is not a lot when you think of all the metrics one could
>     want to keep (e.g. color, FA, curvature, torsion, MD, cluster ID,
>     etc.) in its tractogram file. On top of that, some of these
>     metrics consist of more than one value, for instance the color is
>     in fact composed of three values (i.e. RGB). I find it a bit
>     wasteful to use up multiple tag names only because it is composed
>     of multiple values.
>
>     What I proposed is to encode that number of values in the tag name
>     at save time and take that encoded number into account when
>     loading a TRK file. I have two ways in mind and I would like your
>     opinion.
>
>     *1st approach*
>     A tag name can only have 20 characters/bytes. I would use the last
>     two bytes to encode the number of values associated to a given tag
>     name. The first byte would always be \x00 (EOL) so that TrackVis
>     doesn't try to interpret the last byte which will be set to the
>     number of values (since it is a uint8 the range is [0, 255]). The
>     upside of this approach is that TrackVis is not aware (read
>     doesn't crash) thanks to the \x00 "hack". The downside is that a
>     TrackVis's user will be less aware of what's going on behind the
>     scene but a NiBabel's user won't see a difference since we revert
>     the process when loading the TRK file.
>
>     *2nd approach
>     *This consists of simply adding the number of values in plain
>     ascii text as follows: "property_name-n=3" (or a better naming
>     convention if you have one). The upside is the TrackVis's user
>     will see how many values a given scalar/property is composed of.
>     The downside is it will use up more valuable characters as the
>     number of values gets bigger.
>
>     That said, no matter the approach, there is another small
>     downside. Coming back to the color example, TrackVis will
>     associate the tag name only for the first value (R) and the
>     remaining two values (GB) will have an empty tag name. The
>     following picture shows a TRK file, written using the first
>     approach, that has been loaded in TrackVis. The properties (values
>     per streamline) saved are a color (RGB), the mean curvature and
>     the mean torsion of a streamline.
>
>
>
>
>     For more information see the following PR:
>     https://github.com/nipy/nibabel/pull/391/
>
>     Thanks for your feedback.
>
>     -- 
>     Marc
>
>
>     _______________________________________________
>     Neuroimaging mailing list
>     Neuroimaging at python.org <mailto:Neuroimaging at python.org>
>     https://mail.python.org/mailman/listinfo/neuroimaging
>
>
>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160526/1700a8a4/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 24501 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160526/1700a8a4/attachment-0001.png>

From garyfallidis at gmail.com  Thu May 26 15:31:36 2016
From: garyfallidis at gmail.com (Eleftherios Garyfallidis)
Date: Thu, 26 May 2016 19:31:36 +0000
Subject: [Neuroimaging] NiBabel - Streamlines API - TRK header extension
In-Reply-To: <5747131D.5070602@gmail.com>
References: <5746E704.5040100@gmail.com>
 <CADoUENNvNWUs5-ZvyNbivQFNo_tn295cvFsd6fsqM7eomJQEhQ@mail.gmail.com>
 <5747131D.5070602@gmail.com>
Message-ID: <CAN6JmWPdRwdDrmWzgwoeYxPUARO3CG3_G_LXy-DTuCXYdRWYQQ@mail.gmail.com>

Hi Marc-Alex,

+1 for option 1. I think it is a useful addition and it will not be
disruptive for the developers or users who use the trk format. However, we
should move beyond the Trackvis format as soon as possible. There are many
limitations on the format that make it especially time consuming for
loading and saving large datasets. And the datasets continue grow day by
day.

So, I suggest to move ahead with option 1 and merge the PR asap. After that
let's go back to the previous discussion of what should be the correct
format for modern tractograms and not spend more time on filling up
different corner cases of the Trackvis format.

Thanks!

Cheers,
Eleftherios




On Thu, May 26, 2016 at 11:16 AM Marc-Alexandre C?t? <marc.cote.19 at gmail.com>
wrote:

> On 16-05-26 09:56 AM, Jean-Christophe Houde wrote:
>
> Hi Marc-Alex,
>
> a few small things:
>
> - you say you can save 20 property tags and 20  scalar tags. However, in
> the trk format description you linked, it seems to say only 10 for each.
>
> You are right, it is 10 for each not 20.
>
>
> - For your example file, I guess that means you'd be saving using the hack
> so you know that the first 3 properties written for a track are the RGB
> components, and the follwoing 2 are the mean curvature and mean torsion.
> Just to make sure I understand, that means that you would set n_properties
> to 5, to reflect the total number?
>
> Yes, the n_properties would be set to the total number of values
> associated to properties, in this case 5.
>
>
> If you can share your example file, I'd like to load it in our soft to
> check the compatibility (pretty sure we would be willing to support that).
>
> The file I used is the following:
>
> https://github.com/MarcCote/nibabel/blob/streamlines_api/nibabel/streamlines/tests/data/complex.trk
> Note that this file contains 3 streamlines composed of 1, 2 and 3 random
> points respectively, so they might not be displayable.
>
> Also, regarding your software, can you check that
>
> https://github.com/MarcCote/nibabel/blob/streamlines_api/nibabel/streamlines/tests/data/standard.trk
> is displayed correctly on top of
>
> https://github.com/MarcCote/nibabel/blob/streamlines_api/nibabel/streamlines/tests/data/standard.nii.gz
> as expected (I tested it with TrackVis and dipy.viz and both are
> displaying it correctly).
>
>
> --
> JC
>
> 2016-05-26 8:07 GMT-04:00 Marc <marc.cote.19 at gmail.com>:
>
>> Hi everyone,
>>
>> I'm currently working on an API for streamlines in NiBabel and I would
>> like the feedback of the Nipy community about an unofficial extension for
>> the TRK header.
>>
>> Right now, the TrackVis's file format for streamlines supports conserving
>> additional information for each point and/or streamline. The format refers
>> to that additional information as "scalars" and "properties" respectively
>> (ref: http://www.trackvis.org/docs/?subsect=fileformat). There is a
>> field in the header to keep a tag name for the first 20 scalars and the
>> first 20 properties even though the format supports having up to 2^16
>> scalars and 2^16 properties.
>>
>> First, 20 is not a lot when you think of all the metrics one could want
>> to keep (e.g. color, FA, curvature, torsion, MD, cluster ID, etc.) in its
>> tractogram file. On top of that, some of these metrics consist of more than
>> one value, for instance the color is in fact composed of three values (i.e.
>> RGB). I find it a bit wasteful to use up multiple tag names only because it
>> is composed of multiple values.
>>
>> What I proposed is to encode that number of values in the tag name at
>> save time and take that encoded number into account when loading a TRK
>> file. I have two ways in mind and I would like your opinion.
>>
>> *1st approach*
>> A tag name can only have 20 characters/bytes. I would use the last two
>> bytes to encode the number of values associated to a given tag name. The
>> first byte would always be \x00 (EOL) so that TrackVis doesn't try to
>> interpret the last byte which will be set to the number of values (since it
>> is a uint8 the range is [0, 255]). The upside of this approach is that
>> TrackVis is not aware (read doesn't crash) thanks to the \x00 "hack". The
>> downside is that a TrackVis's user will be less aware of what's going on
>> behind the scene but a NiBabel's user won't see a difference since we
>> revert the process when loading the TRK file.
>>
>>
>> *2nd approach *This consists of simply adding the number of values in
>> plain ascii text as follows: "property_name-n=3" (or a better naming
>> convention if you have one). The upside is the TrackVis's user will see how
>> many values a given scalar/property is composed of. The downside is it will
>> use up more valuable characters as the number of values gets bigger.
>>
>> That said, no matter the approach, there is another small downside.
>> Coming back to the color example, TrackVis will associate the tag name only
>> for the first value (R) and the remaining two values (GB) will have an
>> empty tag name. The following picture shows a TRK file, written using the
>> first approach, that has been loaded in TrackVis. The properties (values
>> per streamline) saved are a color (RGB), the mean curvature and the mean
>> torsion of a streamline.
>>
>>
>>
>>
>> For more information see the following PR:
>> https://github.com/nipy/nibabel/pull/391/
>>
>> Thanks for your feedback.
>>
>> --
>> Marc
>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
>
>
> _______________________________________________
> Neuroimaging mailing listNeuroimaging at python.orghttps://mail.python.org/mailman/listinfo/neuroimaging
>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160526/6a8b95b9/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 24501 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160526/6a8b95b9/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 24501 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160526/6a8b95b9/attachment-0001.png>

From avesani at fbk.eu  Fri May 27 05:07:54 2016
From: avesani at fbk.eu (Paolo Avesani)
Date: Fri, 27 May 2016 11:07:54 +0200
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
In-Reply-To: <CAKADGuq-RxaTq03hNxtTr-uuib-OhKaKOi5tL_2VmQCd2_+oSg@mail.gmail.com>
References: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
 <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>
 <CA+X4w0xfBuK8yy8g=RdXQbF503jtHw1PaNOE8Bs0pH7BQq2Z9A@mail.gmail.com>
 <CAHHh6CbGtSCz7HCarvR6_qHBMX-zrNwKfuY9k6N=svc4BHux_A@mail.gmail.com>
 <CA+X4w0xwbHP9O-eH1WvyaU8ncLb4arZF_j8FvEkKZexXoH3FjA@mail.gmail.com>
 <CAKADGuq6kJRd9q1d9zv7=ggXGAau=3-HxoWXz73_qgxuEUTeKQ@mail.gmail.com>
 <CAHHh6CaU-K5tJeNXdUqdOmy09xYZ8eG36oWPFh1t8jaw7bzVVA@mail.gmail.com>
 <CAKADGuq-RxaTq03hNxtTr-uuib-OhKaKOi5tL_2VmQCd2_+oSg@mail.gmail.com>
Message-ID: <CALXVmhH9Exrfa8=SW280BrviM4BuUH5D6CJ2SQN6X4_z2A1uBQ@mail.gmail.com>

Thanks for your replies.
A brief update.

Samuel is right. I have in mind to manage the multi-shell model from
mrtrix3.
It is clear that up to now it is not viable until the work in progress from
Bago becomes available.

I tried to follow the example suggested by Ariel, focusing on a single
shell csd model (coming from dwi2fod command of mrtrix3),
and setting the sphere as the one used by mrtrix3 (a 300 points). The
"predict" method raised an issue of disalignment of matrix size.

The issue is at line 217 of csdecon.py when a product between
"predict_matrix" and "sh_coeff" takes place.
The "predict matrix" has size (90,45): 90 volumes, 45 sh
The "sh_coeff" has size (145, 174, 145, 45): the xyz dimensions of the
volume, 45 sh

It looks the size of predict matrix is wrong. It should be (145,174,145,90).




On Thu, May 26, 2016 at 4:07 AM, Samuel St-Jean <stjeansam at gmail.com> wrote:

> Indeed, I thought I read the multitissue version was used here, my mistake
> as the question does not directly imply that.
>
> The regular single shell hopefully does the same (at least I can
> personally attest the mrtrix2 version and dipy version give similar fodf up
> to a small rounding factor, but that was before the cholesky decomposition
> step, so now they should behave the same).
> On May 26, 2016 09:37, "Bago" <mrbago at gmail.com> wrote:
>
>> Samuel mrtrix has at least two models implemented. The multi-shell
>> (multi-tissue) model cannot be used with single shell data and the original
>> CSD model cannot be used with multi-shell data.
>>
>> Bago
>>
>> On Wed, May 25, 2016 at 5:13 PM Samuel St-Jean <stjeansam at gmail.com>
>> wrote:
>>
>>> Their wiki explains it, a sqrt(2) to normalize is used in mrtrix3, so
>>> multiplying your coefficients with that and using the mrtrix2 functions
>>> should do it.
>>>
>>> Although dipy only does single shell, so conclude with consideration
>>> that the algorithm is different from mrtrix3. Also, csd is a bad signal
>>> predictor (but good for angle estimation), see the sparc dmri challenge
>>> paper for example.
>>> On May 26, 2016 07:55, "Ariel Rokem" <arokem at gmail.com> wrote:
>>>
>>>>
>>>>
>>>> On Wed, May 25, 2016 at 4:41 PM, Bago <mrbago at gmail.com> wrote:
>>>>
>>>>> I believe they did change their basis (please correct me if I'm wrong
>>>>> but I believe they went from a non-normalized SH basis to a normalized SH
>>>>> basis).
>>>>>
>>>>>
>>>> So they have the same basis as dipy now, but the coefficients appear in
>>>> a different order? That should make life even easier!
>>>>
>>>>
>>>>> Also projecting onto a sphere is one way to _estimate_ the
>>>>> coefficients in a different basis. The cleaner way is to just re-order the
>>>>> coefficients and apply the appropriate scaling. If both basis are
>>>>> normalized (which dipy is) the scaling should be 1 or -1.
>>>>>
>>>>>
>>>> Fair point, but to be just a little bit facetious: given enough points
>>>> on the sphere and knowledge of the target maximal order of the
>>>> coefficients, wouldn't estimating be the same as transforming? Works for
>>>> the FFT, I believe :-)
>>>>
>>>> Bago
>>>>>
>>>>> On Wed, May 25, 2016 at 3:31 PM Ariel Rokem <arokem at gmail.com> wrote:
>>>>>
>>>>>> On Wed, May 25, 2016 at 1:09 PM, Bago <mrbago at gmail.com> wrote:
>>>>>>
>>>>>>> Hi Paolo,
>>>>>>>   mrtrix and dipy define the SH basis slightly differently, so the
>>>>>>> precomputed FOD values need to be adjusted if you want to skip the fit step
>>>>>>> and initialize the Fit object directly. IRC we don't currently have the
>>>>>>> code to do that, but it would be something we'd like to incorporate.
>>>>>>>
>>>>>>> Did they change their basis set when they transitioned to mrtrix3?
>>>>>> We do have these functions:
>>>>>>
>>>>>> https://github.com/nipy/dipy/blob/master/dipy/reconst/shm.py#L852-L923
>>>>>>
>>>>>> That should work with the previous version of mrtrix (mrtrix2?). You
>>>>>> can use these to transform between coefficient sets:
>>>>>>
>>>>>>     sf = sh_to_sf(mrtrix_coeffs, sphere, sh_order,
>>>>>> basis_type='mrtrix')
>>>>>>     dipy_coeffs = sf_to_sh(sf, sphere, sh_order, basis_type=None) #
>>>>>> This defaults to the dipy basis set
>>>>>>
>>>>>> and then use the CSD model object to predict:
>>>>>>
>>>>>>     from dipy.reconst.csdeconv import  ConstrainedSphericalDeconvModel
>>>>>>     csd_model = ConstrainedSphericalDeconvModel(gtab, response,
>>>>>> sh_order=sh_order) # Note: you still need to calculate the response
>>>>>> function!
>>>>>>     pred_signal = csd_model.predict(dipy_coeffs, gtab, S0)
>>>>>>
>>>>>> I think that something like this should work (but I haven't tried it
>>>>>> myself).
>>>>>>
>>>>>>
>>>>>>> I have a WIP version of the multi-shell CSD model on a separate
>>>>>>> branch, I plan on merging it but wasn't intending to get to that for a few
>>>>>>> months. If you'd like to look at before then I can push the branch up to
>>>>>>> github.
>>>>>>>
>>>>>>> Sounds interesting! I'd love to see what you have so far!
>>>>>>
>>>>>> Cheers,
>>>>>>
>>>>>> Ariel
>>>>>>
>>>>>>
>>>>>>> Bago
>>>>>>>
>>>>>>> On Wed, May 25, 2016 at 2:31 AM Paolo Avesani <avesani at fbk.eu>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> I would like to take advantage of the "predict" method of
>>>>>>>> reconstruction models in dipy. The goal is to assess the quality of results.
>>>>>>>>
>>>>>>>> I have already computed the reconstruction models using mrtrix3 and
>>>>>>>> stored the ODF files. For this reason I would need to initialize the csd
>>>>>>>> model by importing the data from ODF stored by mrtrix3.
>>>>>>>>
>>>>>>>> The questions are manifold:
>>>>>>>> - may I initialize the csd model by providing the precomputed
>>>>>>>> values and skipping the "fit" step?
>>>>>>>> - may I import the value of precomputed model from a file stored by
>>>>>>>> mrtrix3?
>>>>>>>> - is the csd model in dipy compliant with the output of multi-shell
>>>>>>>> csd model computed by mrtrix3?
>>>>>>>>
>>>>>>>> I hope my questions and my goal is formulated clearly.
>>>>>>>> Thanks for your support.
>>>>>>>> Paolo
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> Neuroimaging mailing list
>>>>>>>> Neuroimaging at python.org
>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Neuroimaging mailing list
>>>>>>> Neuroimaging at python.org
>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>
>>>>>>> _______________________________________________
>>>>>> Neuroimaging mailing list
>>>>>> Neuroimaging at python.org
>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>


-- 
-------------------------------------------------------
Paolo Avesani
Fondazione Bruno Kessler
via Sommarive 18,
38050 Povo (TN) - I
phone:   +39 0461 314336
fax:        +39 0461 302040
email:     avesani at fbk.eu
web:       avesani.fbk.eu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160527/f1e46da5/attachment-0001.html>

From arokem at gmail.com  Fri May 27 08:23:38 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Fri, 27 May 2016 05:23:38 -0700
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
In-Reply-To: <CALXVmhH9Exrfa8=SW280BrviM4BuUH5D6CJ2SQN6X4_z2A1uBQ@mail.gmail.com>
References: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
 <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>
 <CA+X4w0xfBuK8yy8g=RdXQbF503jtHw1PaNOE8Bs0pH7BQq2Z9A@mail.gmail.com>
 <CAHHh6CbGtSCz7HCarvR6_qHBMX-zrNwKfuY9k6N=svc4BHux_A@mail.gmail.com>
 <CA+X4w0xwbHP9O-eH1WvyaU8ncLb4arZF_j8FvEkKZexXoH3FjA@mail.gmail.com>
 <CAKADGuq6kJRd9q1d9zv7=ggXGAau=3-HxoWXz73_qgxuEUTeKQ@mail.gmail.com>
 <CAHHh6CaU-K5tJeNXdUqdOmy09xYZ8eG36oWPFh1t8jaw7bzVVA@mail.gmail.com>
 <CAKADGuq-RxaTq03hNxtTr-uuib-OhKaKOi5tL_2VmQCd2_+oSg@mail.gmail.com>
 <CALXVmhH9Exrfa8=SW280BrviM4BuUH5D6CJ2SQN6X4_z2A1uBQ@mail.gmail.com>
Message-ID: <CA+X4w0xh2R3hRcxL8SENkegJYqTksZiWjsAq1XhAxbY50N1hhA@mail.gmail.com>

Hi Paolo,

Thanks for the update. Sorry to hear it didn't work as hoped.

On Fri, May 27, 2016 at 2:07 AM, Paolo Avesani <avesani at fbk.eu> wrote:

> Thanks for your replies.
> A brief update.
>
> Samuel is right. I have in mind to manage the multi-shell model from
> mrtrix3.
> It is clear that up to now it is not viable until the work in progress
> from Bago becomes available.
>
> I tried to follow the example suggested by Ariel, focusing on a single
> shell csd model (coming from dwi2fod command of mrtrix3),
> and setting the sphere as the one used by mrtrix3 (a 300 points). The
> "predict" method raised an issue of disalignment of matrix size.
>
> The issue is at line 217 of csdecon.py when a product between
> "predict_matrix" and "sh_coeff" takes place.
> The "predict matrix" has size (90,45): 90 volumes, 45 sh
> The "sh_coeff" has size (145, 174, 145, 45): the xyz dimensions of the
> volume, 45 sh
>
> It looks the size of predict matrix is wrong. It should be
> (145,174,145,90).
>
>
>
Or the predict_matrix should be (45, 90)? Would be useful if you shared a
minimal example that raises this error.

Cheers,

Ariel



>
>
> On Thu, May 26, 2016 at 4:07 AM, Samuel St-Jean <stjeansam at gmail.com>
> wrote:
>
>> Indeed, I thought I read the multitissue version was used here, my
>> mistake as the question does not directly imply that.
>>
>> The regular single shell hopefully does the same (at least I can
>> personally attest the mrtrix2 version and dipy version give similar fodf up
>> to a small rounding factor, but that was before the cholesky decomposition
>> step, so now they should behave the same).
>> On May 26, 2016 09:37, "Bago" <mrbago at gmail.com> wrote:
>>
>>> Samuel mrtrix has at least two models implemented. The multi-shell
>>> (multi-tissue) model cannot be used with single shell data and the original
>>> CSD model cannot be used with multi-shell data.
>>>
>>> Bago
>>>
>>> On Wed, May 25, 2016 at 5:13 PM Samuel St-Jean <stjeansam at gmail.com>
>>> wrote:
>>>
>>>> Their wiki explains it, a sqrt(2) to normalize is used in mrtrix3, so
>>>> multiplying your coefficients with that and using the mrtrix2 functions
>>>> should do it.
>>>>
>>>> Although dipy only does single shell, so conclude with consideration
>>>> that the algorithm is different from mrtrix3. Also, csd is a bad signal
>>>> predictor (but good for angle estimation), see the sparc dmri challenge
>>>> paper for example.
>>>> On May 26, 2016 07:55, "Ariel Rokem" <arokem at gmail.com> wrote:
>>>>
>>>>>
>>>>>
>>>>> On Wed, May 25, 2016 at 4:41 PM, Bago <mrbago at gmail.com> wrote:
>>>>>
>>>>>> I believe they did change their basis (please correct me if I'm wrong
>>>>>> but I believe they went from a non-normalized SH basis to a normalized SH
>>>>>> basis).
>>>>>>
>>>>>>
>>>>> So they have the same basis as dipy now, but the coefficients appear
>>>>> in a different order? That should make life even easier!
>>>>>
>>>>>
>>>>>> Also projecting onto a sphere is one way to _estimate_ the
>>>>>> coefficients in a different basis. The cleaner way is to just re-order the
>>>>>> coefficients and apply the appropriate scaling. If both basis are
>>>>>> normalized (which dipy is) the scaling should be 1 or -1.
>>>>>>
>>>>>>
>>>>> Fair point, but to be just a little bit facetious: given enough points
>>>>> on the sphere and knowledge of the target maximal order of the
>>>>> coefficients, wouldn't estimating be the same as transforming? Works for
>>>>> the FFT, I believe :-)
>>>>>
>>>>> Bago
>>>>>>
>>>>>> On Wed, May 25, 2016 at 3:31 PM Ariel Rokem <arokem at gmail.com> wrote:
>>>>>>
>>>>>>> On Wed, May 25, 2016 at 1:09 PM, Bago <mrbago at gmail.com> wrote:
>>>>>>>
>>>>>>>> Hi Paolo,
>>>>>>>>   mrtrix and dipy define the SH basis slightly differently, so the
>>>>>>>> precomputed FOD values need to be adjusted if you want to skip the fit step
>>>>>>>> and initialize the Fit object directly. IRC we don't currently have the
>>>>>>>> code to do that, but it would be something we'd like to incorporate.
>>>>>>>>
>>>>>>>> Did they change their basis set when they transitioned to mrtrix3?
>>>>>>> We do have these functions:
>>>>>>>
>>>>>>>
>>>>>>> https://github.com/nipy/dipy/blob/master/dipy/reconst/shm.py#L852-L923
>>>>>>>
>>>>>>> That should work with the previous version of mrtrix (mrtrix2?). You
>>>>>>> can use these to transform between coefficient sets:
>>>>>>>
>>>>>>>     sf = sh_to_sf(mrtrix_coeffs, sphere, sh_order,
>>>>>>> basis_type='mrtrix')
>>>>>>>     dipy_coeffs = sf_to_sh(sf, sphere, sh_order, basis_type=None) #
>>>>>>> This defaults to the dipy basis set
>>>>>>>
>>>>>>> and then use the CSD model object to predict:
>>>>>>>
>>>>>>>     from dipy.reconst.csdeconv import
>>>>>>>  ConstrainedSphericalDeconvModel
>>>>>>>     csd_model = ConstrainedSphericalDeconvModel(gtab, response,
>>>>>>> sh_order=sh_order) # Note: you still need to calculate the response
>>>>>>> function!
>>>>>>>     pred_signal = csd_model.predict(dipy_coeffs, gtab, S0)
>>>>>>>
>>>>>>> I think that something like this should work (but I haven't tried it
>>>>>>> myself).
>>>>>>>
>>>>>>>
>>>>>>>> I have a WIP version of the multi-shell CSD model on a separate
>>>>>>>> branch, I plan on merging it but wasn't intending to get to that for a few
>>>>>>>> months. If you'd like to look at before then I can push the branch up to
>>>>>>>> github.
>>>>>>>>
>>>>>>>> Sounds interesting! I'd love to see what you have so far!
>>>>>>>
>>>>>>> Cheers,
>>>>>>>
>>>>>>> Ariel
>>>>>>>
>>>>>>>
>>>>>>>> Bago
>>>>>>>>
>>>>>>>> On Wed, May 25, 2016 at 2:31 AM Paolo Avesani <avesani at fbk.eu>
>>>>>>>> wrote:
>>>>>>>>
>>>>>>>>> I would like to take advantage of the "predict" method of
>>>>>>>>> reconstruction models in dipy. The goal is to assess the quality of results.
>>>>>>>>>
>>>>>>>>> I have already computed the reconstruction models using mrtrix3
>>>>>>>>> and stored the ODF files. For this reason I would need to initialize the
>>>>>>>>> csd model by importing the data from ODF stored by mrtrix3.
>>>>>>>>>
>>>>>>>>> The questions are manifold:
>>>>>>>>> - may I initialize the csd model by providing the precomputed
>>>>>>>>> values and skipping the "fit" step?
>>>>>>>>> - may I import the value of precomputed model from a file stored
>>>>>>>>> by mrtrix3?
>>>>>>>>> - is the csd model in dipy compliant with the output of
>>>>>>>>> multi-shell csd model computed by mrtrix3?
>>>>>>>>>
>>>>>>>>> I hope my questions and my goal is formulated clearly.
>>>>>>>>> Thanks for your support.
>>>>>>>>> Paolo
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> Neuroimaging mailing list
>>>>>>>>> Neuroimaging at python.org
>>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> Neuroimaging mailing list
>>>>>>>> Neuroimaging at python.org
>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>> Neuroimaging mailing list
>>>>>>> Neuroimaging at python.org
>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Neuroimaging mailing list
>>>>>> Neuroimaging at python.org
>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
>
>
> --
> -------------------------------------------------------
> Paolo Avesani
> Fondazione Bruno Kessler
> via Sommarive 18,
> 38050 Povo (TN) - I
> phone:   +39 0461 314336
> fax:        +39 0461 302040
> email:     avesani at fbk.eu
> web:       avesani.fbk.eu
>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160527/b560a676/attachment.html>

From arokem at gmail.com  Fri May 27 11:44:06 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Fri, 27 May 2016 08:44:06 -0700
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
In-Reply-To: <CA+X4w0xh2R3hRcxL8SENkegJYqTksZiWjsAq1XhAxbY50N1hhA@mail.gmail.com>
References: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
 <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>
 <CA+X4w0xfBuK8yy8g=RdXQbF503jtHw1PaNOE8Bs0pH7BQq2Z9A@mail.gmail.com>
 <CAHHh6CbGtSCz7HCarvR6_qHBMX-zrNwKfuY9k6N=svc4BHux_A@mail.gmail.com>
 <CA+X4w0xwbHP9O-eH1WvyaU8ncLb4arZF_j8FvEkKZexXoH3FjA@mail.gmail.com>
 <CAKADGuq6kJRd9q1d9zv7=ggXGAau=3-HxoWXz73_qgxuEUTeKQ@mail.gmail.com>
 <CAHHh6CaU-K5tJeNXdUqdOmy09xYZ8eG36oWPFh1t8jaw7bzVVA@mail.gmail.com>
 <CAKADGuq-RxaTq03hNxtTr-uuib-OhKaKOi5tL_2VmQCd2_+oSg@mail.gmail.com>
 <CALXVmhH9Exrfa8=SW280BrviM4BuUH5D6CJ2SQN6X4_z2A1uBQ@mail.gmail.com>
 <CA+X4w0xh2R3hRcxL8SENkegJYqTksZiWjsAq1XhAxbY50N1hhA@mail.gmail.com>
Message-ID: <CA+X4w0y7w=xxLz07um=-1kohEp5DEutgze7JH4aiXhiHuTiY-g@mail.gmail.com>

Hi again,

On Fri, May 27, 2016 at 5:23 AM, Ariel Rokem <arokem at gmail.com> wrote:

> Hi Paolo,
>
> Thanks for the update. Sorry to hear it didn't work as hoped.
>
> On Fri, May 27, 2016 at 2:07 AM, Paolo Avesani <avesani at fbk.eu> wrote:
>
>> Thanks for your replies.
>> A brief update.
>>
>> Samuel is right. I have in mind to manage the multi-shell model from
>> mrtrix3.
>> It is clear that up to now it is not viable until the work in progress
>> from Bago becomes available.
>>
>> I tried to follow the example suggested by Ariel, focusing on a single
>> shell csd model (coming from dwi2fod command of mrtrix3),
>> and setting the sphere as the one used by mrtrix3 (a 300 points). The
>> "predict" method raised an issue of disalignment of matrix size.
>>
>> The issue is at line 217 of csdecon.py when a product between
>> "predict_matrix" and "sh_coeff" takes place.
>> The "predict matrix" has size (90,45): 90 volumes, 45 sh
>> The "sh_coeff" has size (145, 174, 145, 45): the xyz dimensions of the
>> volume, 45 sh
>>
>> It looks the size of predict matrix is wrong. It should be
>> (145,174,145,90).
>>
>>
>>
> Or the predict_matrix should be (45, 90)? Would be useful if you shared a
> minimal example that raises this error.
>

I was wrong, of course. Does Bago's fix (
https://github.com/nipy/dipy/pull/1062) resolve this for you?

Cheers,

Ariel



> Cheers,
>
> Ariel
>
>
>
>>
>>
>> On Thu, May 26, 2016 at 4:07 AM, Samuel St-Jean <stjeansam at gmail.com>
>> wrote:
>>
>>> Indeed, I thought I read the multitissue version was used here, my
>>> mistake as the question does not directly imply that.
>>>
>>> The regular single shell hopefully does the same (at least I can
>>> personally attest the mrtrix2 version and dipy version give similar fodf up
>>> to a small rounding factor, but that was before the cholesky decomposition
>>> step, so now they should behave the same).
>>> On May 26, 2016 09:37, "Bago" <mrbago at gmail.com> wrote:
>>>
>>>> Samuel mrtrix has at least two models implemented. The multi-shell
>>>> (multi-tissue) model cannot be used with single shell data and the original
>>>> CSD model cannot be used with multi-shell data.
>>>>
>>>> Bago
>>>>
>>>> On Wed, May 25, 2016 at 5:13 PM Samuel St-Jean <stjeansam at gmail.com>
>>>> wrote:
>>>>
>>>>> Their wiki explains it, a sqrt(2) to normalize is used in mrtrix3, so
>>>>> multiplying your coefficients with that and using the mrtrix2 functions
>>>>> should do it.
>>>>>
>>>>> Although dipy only does single shell, so conclude with consideration
>>>>> that the algorithm is different from mrtrix3. Also, csd is a bad signal
>>>>> predictor (but good for angle estimation), see the sparc dmri challenge
>>>>> paper for example.
>>>>> On May 26, 2016 07:55, "Ariel Rokem" <arokem at gmail.com> wrote:
>>>>>
>>>>>>
>>>>>>
>>>>>> On Wed, May 25, 2016 at 4:41 PM, Bago <mrbago at gmail.com> wrote:
>>>>>>
>>>>>>> I believe they did change their basis (please correct me if I'm
>>>>>>> wrong but I believe they went from a non-normalized SH basis to a
>>>>>>> normalized SH basis).
>>>>>>>
>>>>>>>
>>>>>> So they have the same basis as dipy now, but the coefficients appear
>>>>>> in a different order? That should make life even easier!
>>>>>>
>>>>>>
>>>>>>> Also projecting onto a sphere is one way to _estimate_ the
>>>>>>> coefficients in a different basis. The cleaner way is to just re-order the
>>>>>>> coefficients and apply the appropriate scaling. If both basis are
>>>>>>> normalized (which dipy is) the scaling should be 1 or -1.
>>>>>>>
>>>>>>>
>>>>>> Fair point, but to be just a little bit facetious: given enough
>>>>>> points on the sphere and knowledge of the target maximal order of the
>>>>>> coefficients, wouldn't estimating be the same as transforming? Works for
>>>>>> the FFT, I believe :-)
>>>>>>
>>>>>> Bago
>>>>>>>
>>>>>>> On Wed, May 25, 2016 at 3:31 PM Ariel Rokem <arokem at gmail.com>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> On Wed, May 25, 2016 at 1:09 PM, Bago <mrbago at gmail.com> wrote:
>>>>>>>>
>>>>>>>>> Hi Paolo,
>>>>>>>>>   mrtrix and dipy define the SH basis slightly differently, so the
>>>>>>>>> precomputed FOD values need to be adjusted if you want to skip the fit step
>>>>>>>>> and initialize the Fit object directly. IRC we don't currently have the
>>>>>>>>> code to do that, but it would be something we'd like to incorporate.
>>>>>>>>>
>>>>>>>>> Did they change their basis set when they transitioned to mrtrix3?
>>>>>>>> We do have these functions:
>>>>>>>>
>>>>>>>>
>>>>>>>> https://github.com/nipy/dipy/blob/master/dipy/reconst/shm.py#L852-L923
>>>>>>>>
>>>>>>>> That should work with the previous version of mrtrix (mrtrix2?).
>>>>>>>> You can use these to transform between coefficient sets:
>>>>>>>>
>>>>>>>>     sf = sh_to_sf(mrtrix_coeffs, sphere, sh_order,
>>>>>>>> basis_type='mrtrix')
>>>>>>>>     dipy_coeffs = sf_to_sh(sf, sphere, sh_order, basis_type=None) #
>>>>>>>> This defaults to the dipy basis set
>>>>>>>>
>>>>>>>> and then use the CSD model object to predict:
>>>>>>>>
>>>>>>>>     from dipy.reconst.csdeconv import
>>>>>>>>  ConstrainedSphericalDeconvModel
>>>>>>>>     csd_model = ConstrainedSphericalDeconvModel(gtab, response,
>>>>>>>> sh_order=sh_order) # Note: you still need to calculate the response
>>>>>>>> function!
>>>>>>>>     pred_signal = csd_model.predict(dipy_coeffs, gtab, S0)
>>>>>>>>
>>>>>>>> I think that something like this should work (but I haven't tried
>>>>>>>> it myself).
>>>>>>>>
>>>>>>>>
>>>>>>>>> I have a WIP version of the multi-shell CSD model on a separate
>>>>>>>>> branch, I plan on merging it but wasn't intending to get to that for a few
>>>>>>>>> months. If you'd like to look at before then I can push the branch up to
>>>>>>>>> github.
>>>>>>>>>
>>>>>>>>> Sounds interesting! I'd love to see what you have so far!
>>>>>>>>
>>>>>>>> Cheers,
>>>>>>>>
>>>>>>>> Ariel
>>>>>>>>
>>>>>>>>
>>>>>>>>> Bago
>>>>>>>>>
>>>>>>>>> On Wed, May 25, 2016 at 2:31 AM Paolo Avesani <avesani at fbk.eu>
>>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>>> I would like to take advantage of the "predict" method of
>>>>>>>>>> reconstruction models in dipy. The goal is to assess the quality of results.
>>>>>>>>>>
>>>>>>>>>> I have already computed the reconstruction models using mrtrix3
>>>>>>>>>> and stored the ODF files. For this reason I would need to initialize the
>>>>>>>>>> csd model by importing the data from ODF stored by mrtrix3.
>>>>>>>>>>
>>>>>>>>>> The questions are manifold:
>>>>>>>>>> - may I initialize the csd model by providing the precomputed
>>>>>>>>>> values and skipping the "fit" step?
>>>>>>>>>> - may I import the value of precomputed model from a file stored
>>>>>>>>>> by mrtrix3?
>>>>>>>>>> - is the csd model in dipy compliant with the output of
>>>>>>>>>> multi-shell csd model computed by mrtrix3?
>>>>>>>>>>
>>>>>>>>>> I hope my questions and my goal is formulated clearly.
>>>>>>>>>> Thanks for your support.
>>>>>>>>>> Paolo
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> Neuroimaging mailing list
>>>>>>>>>> Neuroimaging at python.org
>>>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> Neuroimaging mailing list
>>>>>>>>> Neuroimaging at python.org
>>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>> Neuroimaging mailing list
>>>>>>>> Neuroimaging at python.org
>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Neuroimaging mailing list
>>>>>>> Neuroimaging at python.org
>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Neuroimaging mailing list
>>>>>> Neuroimaging at python.org
>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>
>>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>>
>>
>>
>> --
>> -------------------------------------------------------
>> Paolo Avesani
>> Fondazione Bruno Kessler
>> via Sommarive 18,
>> 38050 Povo (TN) - I
>> phone:   +39 0461 314336
>> fax:        +39 0461 302040
>> email:     avesani at fbk.eu
>> web:       avesani.fbk.eu
>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160527/0eac3bbc/attachment.html>

From mrbago at gmail.com  Fri May 27 12:31:40 2016
From: mrbago at gmail.com (Bago)
Date: Fri, 27 May 2016 16:31:40 +0000
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
In-Reply-To: <CA+X4w0y7w=xxLz07um=-1kohEp5DEutgze7JH4aiXhiHuTiY-g@mail.gmail.com>
References: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
 <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>
 <CA+X4w0xfBuK8yy8g=RdXQbF503jtHw1PaNOE8Bs0pH7BQq2Z9A@mail.gmail.com>
 <CAHHh6CbGtSCz7HCarvR6_qHBMX-zrNwKfuY9k6N=svc4BHux_A@mail.gmail.com>
 <CA+X4w0xwbHP9O-eH1WvyaU8ncLb4arZF_j8FvEkKZexXoH3FjA@mail.gmail.com>
 <CAKADGuq6kJRd9q1d9zv7=ggXGAau=3-HxoWXz73_qgxuEUTeKQ@mail.gmail.com>
 <CAHHh6CaU-K5tJeNXdUqdOmy09xYZ8eG36oWPFh1t8jaw7bzVVA@mail.gmail.com>
 <CAKADGuq-RxaTq03hNxtTr-uuib-OhKaKOi5tL_2VmQCd2_+oSg@mail.gmail.com>
 <CALXVmhH9Exrfa8=SW280BrviM4BuUH5D6CJ2SQN6X4_z2A1uBQ@mail.gmail.com>
 <CA+X4w0xh2R3hRcxL8SENkegJYqTksZiWjsAq1XhAxbY50N1hhA@mail.gmail.com>
 <CA+X4w0y7w=xxLz07um=-1kohEp5DEutgze7JH4aiXhiHuTiY-g@mail.gmail.com>
Message-ID: <CAHHh6CZr0KWLdVZjMB4Z=-gKFyOBmjVDbgoiXFq0qxyvjFNnVA@mail.gmail.com>

You, were spot on, we just needed a transpose :).

On Fri, May 27, 2016, 8:44 AM Ariel Rokem <arokem at gmail.com> wrote:

> Hi again,
>
> On Fri, May 27, 2016 at 5:23 AM, Ariel Rokem <arokem at gmail.com> wrote:
>
>> Hi Paolo,
>>
>> Thanks for the update. Sorry to hear it didn't work as hoped.
>>
>> On Fri, May 27, 2016 at 2:07 AM, Paolo Avesani <avesani at fbk.eu> wrote:
>>
>>> Thanks for your replies.
>>> A brief update.
>>>
>>> Samuel is right. I have in mind to manage the multi-shell model from
>>> mrtrix3.
>>> It is clear that up to now it is not viable until the work in progress
>>> from Bago becomes available.
>>>
>>> I tried to follow the example suggested by Ariel, focusing on a single
>>> shell csd model (coming from dwi2fod command of mrtrix3),
>>> and setting the sphere as the one used by mrtrix3 (a 300 points). The
>>> "predict" method raised an issue of disalignment of matrix size.
>>>
>>> The issue is at line 217 of csdecon.py when a product between
>>> "predict_matrix" and "sh_coeff" takes place.
>>> The "predict matrix" has size (90,45): 90 volumes, 45 sh
>>> The "sh_coeff" has size (145, 174, 145, 45): the xyz dimensions of the
>>> volume, 45 sh
>>>
>>> It looks the size of predict matrix is wrong. It should be
>>> (145,174,145,90).
>>>
>>>
>>>
>> Or the predict_matrix should be (45, 90)? Would be useful if you shared a
>> minimal example that raises this error.
>>
>
> I was wrong, of course. Does Bago's fix (
> https://github.com/nipy/dipy/pull/1062) resolve this for you?
>
> Cheers,
>
> Ariel
>
>
>
>> Cheers,
>>
>> Ariel
>>
>>
>>
>>>
>>>
>>> On Thu, May 26, 2016 at 4:07 AM, Samuel St-Jean <stjeansam at gmail.com>
>>> wrote:
>>>
>>>> Indeed, I thought I read the multitissue version was used here, my
>>>> mistake as the question does not directly imply that.
>>>>
>>>> The regular single shell hopefully does the same (at least I can
>>>> personally attest the mrtrix2 version and dipy version give similar fodf up
>>>> to a small rounding factor, but that was before the cholesky decomposition
>>>> step, so now they should behave the same).
>>>> On May 26, 2016 09:37, "Bago" <mrbago at gmail.com> wrote:
>>>>
>>>>> Samuel mrtrix has at least two models implemented. The multi-shell
>>>>> (multi-tissue) model cannot be used with single shell data and the original
>>>>> CSD model cannot be used with multi-shell data.
>>>>>
>>>>> Bago
>>>>>
>>>>> On Wed, May 25, 2016 at 5:13 PM Samuel St-Jean <stjeansam at gmail.com>
>>>>> wrote:
>>>>>
>>>>>> Their wiki explains it, a sqrt(2) to normalize is used in mrtrix3, so
>>>>>> multiplying your coefficients with that and using the mrtrix2 functions
>>>>>> should do it.
>>>>>>
>>>>>> Although dipy only does single shell, so conclude with consideration
>>>>>> that the algorithm is different from mrtrix3. Also, csd is a bad signal
>>>>>> predictor (but good for angle estimation), see the sparc dmri challenge
>>>>>> paper for example.
>>>>>> On May 26, 2016 07:55, "Ariel Rokem" <arokem at gmail.com> wrote:
>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On Wed, May 25, 2016 at 4:41 PM, Bago <mrbago at gmail.com> wrote:
>>>>>>>
>>>>>>>> I believe they did change their basis (please correct me if I'm
>>>>>>>> wrong but I believe they went from a non-normalized SH basis to a
>>>>>>>> normalized SH basis).
>>>>>>>>
>>>>>>>>
>>>>>>> So they have the same basis as dipy now, but the coefficients appear
>>>>>>> in a different order? That should make life even easier!
>>>>>>>
>>>>>>>
>>>>>>>> Also projecting onto a sphere is one way to _estimate_ the
>>>>>>>> coefficients in a different basis. The cleaner way is to just re-order the
>>>>>>>> coefficients and apply the appropriate scaling. If both basis are
>>>>>>>> normalized (which dipy is) the scaling should be 1 or -1.
>>>>>>>>
>>>>>>>>
>>>>>>> Fair point, but to be just a little bit facetious: given enough
>>>>>>> points on the sphere and knowledge of the target maximal order of the
>>>>>>> coefficients, wouldn't estimating be the same as transforming? Works for
>>>>>>> the FFT, I believe :-)
>>>>>>>
>>>>>>> Bago
>>>>>>>>
>>>>>>>> On Wed, May 25, 2016 at 3:31 PM Ariel Rokem <arokem at gmail.com>
>>>>>>>> wrote:
>>>>>>>>
>>>>>>>>> On Wed, May 25, 2016 at 1:09 PM, Bago <mrbago at gmail.com> wrote:
>>>>>>>>>
>>>>>>>>>> Hi Paolo,
>>>>>>>>>>   mrtrix and dipy define the SH basis slightly differently, so
>>>>>>>>>> the precomputed FOD values need to be adjusted if you want to skip the fit
>>>>>>>>>> step and initialize the Fit object directly. IRC we don't currently have
>>>>>>>>>> the code to do that, but it would be something we'd like to incorporate.
>>>>>>>>>>
>>>>>>>>>> Did they change their basis set when they transitioned to
>>>>>>>>> mrtrix3? We do have these functions:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> https://github.com/nipy/dipy/blob/master/dipy/reconst/shm.py#L852-L923
>>>>>>>>>
>>>>>>>>> That should work with the previous version of mrtrix (mrtrix2?).
>>>>>>>>> You can use these to transform between coefficient sets:
>>>>>>>>>
>>>>>>>>>     sf = sh_to_sf(mrtrix_coeffs, sphere, sh_order,
>>>>>>>>> basis_type='mrtrix')
>>>>>>>>>     dipy_coeffs = sf_to_sh(sf, sphere, sh_order, basis_type=None)
>>>>>>>>> # This defaults to the dipy basis set
>>>>>>>>>
>>>>>>>>> and then use the CSD model object to predict:
>>>>>>>>>
>>>>>>>>>     from dipy.reconst.csdeconv import
>>>>>>>>>  ConstrainedSphericalDeconvModel
>>>>>>>>>     csd_model = ConstrainedSphericalDeconvModel(gtab, response,
>>>>>>>>> sh_order=sh_order) # Note: you still need to calculate the response
>>>>>>>>> function!
>>>>>>>>>     pred_signal = csd_model.predict(dipy_coeffs, gtab, S0)
>>>>>>>>>
>>>>>>>>> I think that something like this should work (but I haven't tried
>>>>>>>>> it myself).
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>> I have a WIP version of the multi-shell CSD model on a separate
>>>>>>>>>> branch, I plan on merging it but wasn't intending to get to that for a few
>>>>>>>>>> months. If you'd like to look at before then I can push the branch up to
>>>>>>>>>> github.
>>>>>>>>>>
>>>>>>>>>> Sounds interesting! I'd love to see what you have so far!
>>>>>>>>>
>>>>>>>>> Cheers,
>>>>>>>>>
>>>>>>>>> Ariel
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>> Bago
>>>>>>>>>>
>>>>>>>>>> On Wed, May 25, 2016 at 2:31 AM Paolo Avesani <avesani at fbk.eu>
>>>>>>>>>> wrote:
>>>>>>>>>>
>>>>>>>>>>> I would like to take advantage of the "predict" method of
>>>>>>>>>>> reconstruction models in dipy. The goal is to assess the quality of results.
>>>>>>>>>>>
>>>>>>>>>>> I have already computed the reconstruction models using mrtrix3
>>>>>>>>>>> and stored the ODF files. For this reason I would need to initialize the
>>>>>>>>>>> csd model by importing the data from ODF stored by mrtrix3.
>>>>>>>>>>>
>>>>>>>>>>> The questions are manifold:
>>>>>>>>>>> - may I initialize the csd model by providing the precomputed
>>>>>>>>>>> values and skipping the "fit" step?
>>>>>>>>>>> - may I import the value of precomputed model from a file stored
>>>>>>>>>>> by mrtrix3?
>>>>>>>>>>> - is the csd model in dipy compliant with the output of
>>>>>>>>>>> multi-shell csd model computed by mrtrix3?
>>>>>>>>>>>
>>>>>>>>>>> I hope my questions and my goal is formulated clearly.
>>>>>>>>>>> Thanks for your support.
>>>>>>>>>>> Paolo
>>>>>>>>>>>
>>>>>>>>>>> _______________________________________________
>>>>>>>>>>> Neuroimaging mailing list
>>>>>>>>>>> Neuroimaging at python.org
>>>>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> Neuroimaging mailing list
>>>>>>>>>> Neuroimaging at python.org
>>>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>> Neuroimaging mailing list
>>>>>>>>> Neuroimaging at python.org
>>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> Neuroimaging mailing list
>>>>>>>> Neuroimaging at python.org
>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Neuroimaging mailing list
>>>>>>> Neuroimaging at python.org
>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>
>>>>>>> _______________________________________________
>>>>>> Neuroimaging mailing list
>>>>>> Neuroimaging at python.org
>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>>
>>>
>>>
>>> --
>>> -------------------------------------------------------
>>> Paolo Avesani
>>> Fondazione Bruno Kessler
>>> via Sommarive 18,
>>> 38050 Povo (TN) - I
>>> phone:   +39 0461 314336
>>> fax:        +39 0461 302040
>>> email:     avesani at fbk.eu
>>> web:       avesani.fbk.eu
>>>
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>>
>> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160527/ccf19905/attachment.html>

From arokem at gmail.com  Fri May 27 12:34:22 2016
From: arokem at gmail.com (Ariel Rokem)
Date: Fri, 27 May 2016 09:34:22 -0700
Subject: [Neuroimaging] [dipy] Import csd model precomputed by mrtrix
In-Reply-To: <CAHHh6CZr0KWLdVZjMB4Z=-gKFyOBmjVDbgoiXFq0qxyvjFNnVA@mail.gmail.com>
References: <CALXVmhFvFKsbonQQsuTTq=KBNBYTJcPnzLaU+TjbPYaMk+Gkhw@mail.gmail.com>
 <CAHHh6Cayudb13RYt5ihKPc3HE6w=ra+HU+XYNdWuwea02aFzxw@mail.gmail.com>
 <CA+X4w0xfBuK8yy8g=RdXQbF503jtHw1PaNOE8Bs0pH7BQq2Z9A@mail.gmail.com>
 <CAHHh6CbGtSCz7HCarvR6_qHBMX-zrNwKfuY9k6N=svc4BHux_A@mail.gmail.com>
 <CA+X4w0xwbHP9O-eH1WvyaU8ncLb4arZF_j8FvEkKZexXoH3FjA@mail.gmail.com>
 <CAKADGuq6kJRd9q1d9zv7=ggXGAau=3-HxoWXz73_qgxuEUTeKQ@mail.gmail.com>
 <CAHHh6CaU-K5tJeNXdUqdOmy09xYZ8eG36oWPFh1t8jaw7bzVVA@mail.gmail.com>
 <CAKADGuq-RxaTq03hNxtTr-uuib-OhKaKOi5tL_2VmQCd2_+oSg@mail.gmail.com>
 <CALXVmhH9Exrfa8=SW280BrviM4BuUH5D6CJ2SQN6X4_z2A1uBQ@mail.gmail.com>
 <CA+X4w0xh2R3hRcxL8SENkegJYqTksZiWjsAq1XhAxbY50N1hhA@mail.gmail.com>
 <CA+X4w0y7w=xxLz07um=-1kohEp5DEutgze7JH4aiXhiHuTiY-g@mail.gmail.com>
 <CAHHh6CZr0KWLdVZjMB4Z=-gKFyOBmjVDbgoiXFq0qxyvjFNnVA@mail.gmail.com>
Message-ID: <CA+X4w0yUkb3LA-viV0bVh+bpe97Xexdm5xF5e603odgJJwMq-w@mail.gmail.com>

On Fri, May 27, 2016 at 9:31 AM, Bago <mrbago at gmail.com> wrote:

> You, were spot on, we just needed a transpose :).
>

Ah yes! of course I was :-)




> On Fri, May 27, 2016, 8:44 AM Ariel Rokem <arokem at gmail.com> wrote:
>
>> Hi again,
>>
>> On Fri, May 27, 2016 at 5:23 AM, Ariel Rokem <arokem at gmail.com> wrote:
>>
>>> Hi Paolo,
>>>
>>> Thanks for the update. Sorry to hear it didn't work as hoped.
>>>
>>> On Fri, May 27, 2016 at 2:07 AM, Paolo Avesani <avesani at fbk.eu> wrote:
>>>
>>>> Thanks for your replies.
>>>> A brief update.
>>>>
>>>> Samuel is right. I have in mind to manage the multi-shell model from
>>>> mrtrix3.
>>>> It is clear that up to now it is not viable until the work in progress
>>>> from Bago becomes available.
>>>>
>>>> I tried to follow the example suggested by Ariel, focusing on a single
>>>> shell csd model (coming from dwi2fod command of mrtrix3),
>>>> and setting the sphere as the one used by mrtrix3 (a 300 points). The
>>>> "predict" method raised an issue of disalignment of matrix size.
>>>>
>>>> The issue is at line 217 of csdecon.py when a product between
>>>> "predict_matrix" and "sh_coeff" takes place.
>>>> The "predict matrix" has size (90,45): 90 volumes, 45 sh
>>>> The "sh_coeff" has size (145, 174, 145, 45): the xyz dimensions of the
>>>> volume, 45 sh
>>>>
>>>> It looks the size of predict matrix is wrong. It should be
>>>> (145,174,145,90).
>>>>
>>>>
>>>>
>>> Or the predict_matrix should be (45, 90)? Would be useful if you shared
>>> a minimal example that raises this error.
>>>
>>
>> I was wrong, of course. Does Bago's fix (
>> https://github.com/nipy/dipy/pull/1062) resolve this for you?
>>
>> Cheers,
>>
>> Ariel
>>
>>
>>
>>> Cheers,
>>>
>>> Ariel
>>>
>>>
>>>
>>>>
>>>>
>>>> On Thu, May 26, 2016 at 4:07 AM, Samuel St-Jean <stjeansam at gmail.com>
>>>> wrote:
>>>>
>>>>> Indeed, I thought I read the multitissue version was used here, my
>>>>> mistake as the question does not directly imply that.
>>>>>
>>>>> The regular single shell hopefully does the same (at least I can
>>>>> personally attest the mrtrix2 version and dipy version give similar fodf up
>>>>> to a small rounding factor, but that was before the cholesky decomposition
>>>>> step, so now they should behave the same).
>>>>> On May 26, 2016 09:37, "Bago" <mrbago at gmail.com> wrote:
>>>>>
>>>>>> Samuel mrtrix has at least two models implemented. The multi-shell
>>>>>> (multi-tissue) model cannot be used with single shell data and the original
>>>>>> CSD model cannot be used with multi-shell data.
>>>>>>
>>>>>> Bago
>>>>>>
>>>>>> On Wed, May 25, 2016 at 5:13 PM Samuel St-Jean <stjeansam at gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>> Their wiki explains it, a sqrt(2) to normalize is used in mrtrix3,
>>>>>>> so multiplying your coefficients with that and using the mrtrix2 functions
>>>>>>> should do it.
>>>>>>>
>>>>>>> Although dipy only does single shell, so conclude with consideration
>>>>>>> that the algorithm is different from mrtrix3. Also, csd is a bad signal
>>>>>>> predictor (but good for angle estimation), see the sparc dmri challenge
>>>>>>> paper for example.
>>>>>>> On May 26, 2016 07:55, "Ariel Rokem" <arokem at gmail.com> wrote:
>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> On Wed, May 25, 2016 at 4:41 PM, Bago <mrbago at gmail.com> wrote:
>>>>>>>>
>>>>>>>>> I believe they did change their basis (please correct me if I'm
>>>>>>>>> wrong but I believe they went from a non-normalized SH basis to a
>>>>>>>>> normalized SH basis).
>>>>>>>>>
>>>>>>>>>
>>>>>>>> So they have the same basis as dipy now, but the coefficients
>>>>>>>> appear in a different order? That should make life even easier!
>>>>>>>>
>>>>>>>>
>>>>>>>>> Also projecting onto a sphere is one way to _estimate_ the
>>>>>>>>> coefficients in a different basis. The cleaner way is to just re-order the
>>>>>>>>> coefficients and apply the appropriate scaling. If both basis are
>>>>>>>>> normalized (which dipy is) the scaling should be 1 or -1.
>>>>>>>>>
>>>>>>>>>
>>>>>>>> Fair point, but to be just a little bit facetious: given enough
>>>>>>>> points on the sphere and knowledge of the target maximal order of the
>>>>>>>> coefficients, wouldn't estimating be the same as transforming? Works for
>>>>>>>> the FFT, I believe :-)
>>>>>>>>
>>>>>>>> Bago
>>>>>>>>>
>>>>>>>>> On Wed, May 25, 2016 at 3:31 PM Ariel Rokem <arokem at gmail.com>
>>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>>> On Wed, May 25, 2016 at 1:09 PM, Bago <mrbago at gmail.com> wrote:
>>>>>>>>>>
>>>>>>>>>>> Hi Paolo,
>>>>>>>>>>>   mrtrix and dipy define the SH basis slightly differently, so
>>>>>>>>>>> the precomputed FOD values need to be adjusted if you want to skip the fit
>>>>>>>>>>> step and initialize the Fit object directly. IRC we don't currently have
>>>>>>>>>>> the code to do that, but it would be something we'd like to incorporate.
>>>>>>>>>>>
>>>>>>>>>>> Did they change their basis set when they transitioned to
>>>>>>>>>> mrtrix3? We do have these functions:
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> https://github.com/nipy/dipy/blob/master/dipy/reconst/shm.py#L852-L923
>>>>>>>>>>
>>>>>>>>>> That should work with the previous version of mrtrix (mrtrix2?).
>>>>>>>>>> You can use these to transform between coefficient sets:
>>>>>>>>>>
>>>>>>>>>>     sf = sh_to_sf(mrtrix_coeffs, sphere, sh_order,
>>>>>>>>>> basis_type='mrtrix')
>>>>>>>>>>     dipy_coeffs = sf_to_sh(sf, sphere, sh_order, basis_type=None)
>>>>>>>>>> # This defaults to the dipy basis set
>>>>>>>>>>
>>>>>>>>>> and then use the CSD model object to predict:
>>>>>>>>>>
>>>>>>>>>>     from dipy.reconst.csdeconv import
>>>>>>>>>>  ConstrainedSphericalDeconvModel
>>>>>>>>>>     csd_model = ConstrainedSphericalDeconvModel(gtab, response,
>>>>>>>>>> sh_order=sh_order) # Note: you still need to calculate the response
>>>>>>>>>> function!
>>>>>>>>>>     pred_signal = csd_model.predict(dipy_coeffs, gtab, S0)
>>>>>>>>>>
>>>>>>>>>> I think that something like this should work (but I haven't tried
>>>>>>>>>> it myself).
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>> I have a WIP version of the multi-shell CSD model on a separate
>>>>>>>>>>> branch, I plan on merging it but wasn't intending to get to that for a few
>>>>>>>>>>> months. If you'd like to look at before then I can push the branch up to
>>>>>>>>>>> github.
>>>>>>>>>>>
>>>>>>>>>>> Sounds interesting! I'd love to see what you have so far!
>>>>>>>>>>
>>>>>>>>>> Cheers,
>>>>>>>>>>
>>>>>>>>>> Ariel
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>> Bago
>>>>>>>>>>>
>>>>>>>>>>> On Wed, May 25, 2016 at 2:31 AM Paolo Avesani <avesani at fbk.eu>
>>>>>>>>>>> wrote:
>>>>>>>>>>>
>>>>>>>>>>>> I would like to take advantage of the "predict" method of
>>>>>>>>>>>> reconstruction models in dipy. The goal is to assess the quality of results.
>>>>>>>>>>>>
>>>>>>>>>>>> I have already computed the reconstruction models using mrtrix3
>>>>>>>>>>>> and stored the ODF files. For this reason I would need to initialize the
>>>>>>>>>>>> csd model by importing the data from ODF stored by mrtrix3.
>>>>>>>>>>>>
>>>>>>>>>>>> The questions are manifold:
>>>>>>>>>>>> - may I initialize the csd model by providing the precomputed
>>>>>>>>>>>> values and skipping the "fit" step?
>>>>>>>>>>>> - may I import the value of precomputed model from a file
>>>>>>>>>>>> stored by mrtrix3?
>>>>>>>>>>>> - is the csd model in dipy compliant with the output of
>>>>>>>>>>>> multi-shell csd model computed by mrtrix3?
>>>>>>>>>>>>
>>>>>>>>>>>> I hope my questions and my goal is formulated clearly.
>>>>>>>>>>>> Thanks for your support.
>>>>>>>>>>>> Paolo
>>>>>>>>>>>>
>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>> Neuroimaging mailing list
>>>>>>>>>>>> Neuroimaging at python.org
>>>>>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> _______________________________________________
>>>>>>>>>>> Neuroimaging mailing list
>>>>>>>>>>> Neuroimaging at python.org
>>>>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>>>>
>>>>>>>>>>> _______________________________________________
>>>>>>>>>> Neuroimaging mailing list
>>>>>>>>>> Neuroimaging at python.org
>>>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> Neuroimaging mailing list
>>>>>>>>> Neuroimaging at python.org
>>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> Neuroimaging mailing list
>>>>>>>> Neuroimaging at python.org
>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>> Neuroimaging mailing list
>>>>>>> Neuroimaging at python.org
>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Neuroimaging mailing list
>>>>>> Neuroimaging at python.org
>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>
>>>>>>
>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> -------------------------------------------------------
>>>> Paolo Avesani
>>>> Fondazione Bruno Kessler
>>>> via Sommarive 18,
>>>> 38050 Povo (TN) - I
>>>> phone:   +39 0461 314336
>>>> fax:        +39 0461 302040
>>>> email:     avesani at fbk.eu
>>>> web:       avesani.fbk.eu
>>>>
>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>>
>>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160527/2a6b9a87/attachment.html>

From jhlegarreta at vicomtech.org  Mon May 30 17:50:33 2016
From: jhlegarreta at vicomtech.org (Jon Haitz Legarreta)
Date: Mon, 30 May 2016 23:50:33 +0200
Subject: [Neuroimaging] Fwd: [dipy] Issues trying to install dipy
In-Reply-To: <CA+X4w0z=1qjsbLcknNSjMZOkAjrSxhE=8B23E4+_YJPjcJTVrw@mail.gmail.com>
References: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>
 <CAAKhGh9-emozuiqQv2KpY8hghks-1b-bid1PQufJ7YUwDffoDQ@mail.gmail.com>
 <CAH6Pt5rVree2gyERp3DSXVmC_dS4j4gQ66YJpouO62W4JpvX=g@mail.gmail.com>
 <CA+X4w0zbG_+5FQx8ALCvij54y+p_4WyfpwfD3LyigWoxAqH66A@mail.gmail.com>
 <CAAKhGh_7yUqHe7EVVmpNvNSbZuv1029CfjvK8Vb3Od57FRssEg@mail.gmail.com>
 <CA+X4w0z=1qjsbLcknNSjMZOkAjrSxhE=8B23E4+_YJPjcJTVrw@mail.gmail.com>
Message-ID: <CAAKhGh800SY0ooqJe9_X1BVnOzbVqjGVtaYOHU-zHGLS1n+1jQ@mail.gmail.com>

Dear Ariel,
thanks for your suggestion.

The patches in the link seem to help a little bit, but the process seems
still to be unsuccessful: the MinGW gcc complains with the message:
gcc: error: /arch:SSE2: No such file or directory

Attached is the new log.

Again, googling was not of much help. I got bits and parts of related
errors, but have no clear picture of the issue.

Any suggestion is appreciated.

JON HAITZ



On 24 May 2016 at 01:08, Ariel Rokem <arokem at gmail.com> wrote:

>
>
> On Mon, May 23, 2016 at 3:19 PM, Jon Haitz Legarreta <
> jhlegarreta at vicomtech.org> wrote:
>
>> Hi there,
>> thank you Matthew and Ariel.
>>
>> The link pointed by Ariel does not seem to be a solution; after having
>> installed MinGW, as suggested in the link and although I'm aware it might
>> be unnecessary, the Anaconda3 powershell still yields a similar error, now
>> pointing to MSVC (which I do not have on my system):
>>
>>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 157, in
>> __init__
>>     self.dll_libraries = get_msvcr()
>>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 86, in
>> get_msvcr
>>     raise ValueError("Unknown MS Compiler version %s " % msc_ver)
>> ValueError: Unknown MS Compiler version 1900
>>
>> Looks like maybe you ran into this corner case?
>
> http://stackoverflow.com/a/34427014/3532933
>
>
>
>
>
>> I'll try to investigate further, and will let you know.
>>
>> Kind regards,
>> JON HAITZ
>>
>>
>>
>>
>>
>>
>> On 21 May 2016 at 16:55, Ariel Rokem <arokem at gmail.com> wrote:
>>
>>> Hi Jon and Matthew,
>>>
>>>
>>> On Sat, May 21, 2016 at 7:30 AM, Matthew Brett <matthew.brett at gmail.com>
>>> wrote:
>>>
>>>> Hi,
>>>>
>>>> On Sat, May 21, 2016 at 9:18 AM, Jon Haitz Legarreta
>>>> <jhlegarreta at vicomtech.org> wrote:
>>>> > Hi there,
>>>> > has anybody experienced the issue below?
>>>> >
>>>> > Thanks,
>>>> > JON HAITZ
>>>> >
>>>> >
>>>> >
>>>> >
>>>> > ---------- Forwarded message ----------
>>>> > From: Jon Haitz Legarreta <jhlegarreta at vicomtech.org>
>>>> > Date: 18 May 2016 at 19:10
>>>> > Subject: [dipy] Issues trying to install dipy
>>>> > To: neuroimaging at python.org
>>>> >
>>>> >
>>>> > Hi there,
>>>> > I'm a newbie to dipy.
>>>> >
>>>> > I was trying to follow the instructions in [1] to have dipy installed
>>>> from
>>>> > the source code, so that I could execute the dipy examples.
>>>> >
>>>> > I'm using Windows 10 and Anaconda 3.
>>>> >
>>>> > When trying to execute
>>>> > python setup.py develop
>>>> >
>>>> > the Anaconda prompt yields an error that says in the end:
>>>> > File: "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126,
>>>> __init__
>>>> > if self.ld_version >= "2.10.90":
>>>> > TypeError: unorderable types: NoneType() >= str()
>>>> >
>>>> > I've been googling for a solution without success.
>>>> >
>>>> > I don't know whether this looks like Anaconda3 is trying to use cygwin
>>>> > instead of mingw32, and whether that is the root cause.
>>>> >
>>>> > In either case, does anyone know how to solve the issue?
>>>> >
>>>> > Attached is the trace (it's short) of the error if this is of any
>>>> help.
>>>> >
>>>> >
>>>> > Thank you,
>>>> > JON HAITZ
>>>> >
>>>> >
>>>> > [1] http://nipy.org/dipy/installation.html#install-source-nix
>>>>
>>>> I'm sorry, I'm afraid I don't personally use Anaconda, so I have no
>>>> experience of fixing compilation errors on Anaconda.   Ariel - have
>>>> you come across this?
>>>>
>>>
>>> And I don't personally use Windows...
>>>
>>> Might this be helpful:
>>>
>>>
>>> http://stackoverflow.com/questions/24683305/python-cant-install-packages-typeerror-unorderable-types-nonetype-str
>>>
>>> It seems like it could be related, though it's all Greek to me.
>>>
>>> Cheers,
>>>
>>> Ariel
>>>
>>>
>>>> You could also try on the anaconda support channels (issues, mailing
>>>> list) - it may well be a general problem rather than one specific to
>>>> dipy,
>>>>
>>>> Best,
>>>>
>>>> Matthew
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160530/70110903/attachment.html>
-------------- next part --------------
running develop
running egg_info
writing dipy.egg-info\PKG-INFO
writing requirements to dipy.egg-info\requires.txt
writing dependency_links to dipy.egg-info\dependency_links.txt
writing top-level names to dipy.egg-info\top_level.txt
C:\Anaconda3\lib\site-packages\setuptools-19.6.2-py3.5.egg\setuptools\dist.py:285: UserWarning: Normalizing '0.12.0dev' to '0.12.0.dev0'
package init file 'dipy\data\tests\__init__.py' not found (or not a regular file)
reading manifest file 'dipy.egg-info\SOURCES.txt'
reading manifest template 'MANIFEST.in'
warning: no files found matching 'Changelog'
warning: no files found matching 'TODO'
writing manifest file 'dipy.egg-info\SOURCES.txt'
running build_ext
C:\MinGW\bin\gcc.exe -mdll -O -Wall -IC:\Anaconda3\include -IC:\Anaconda3\include -c test.c -o test.o /arch:SSE2
gcc: error: /arch:SSE2: No such file or directory
Flags ['/arch:SSE2'] omitted because of compile or link error
C:\MinGW\bin\gcc.exe -mdll -O -Wall -IC:\Anaconda3\include -IC:\Anaconda3\include -c test.c -o test.o -msse2 -mfpmath=sse
C:\MinGW\bin\gcc.exe -shared -s test.o -LC:\Anaconda3\libs -LC:\Anaconda3\PCbuild\amd64 -lvcruntime140 -o testlib.dll
c:/mingw/bin/../lib/gcc/mingw32/4.9.3/../../../../mingw32/bin/ld.exe: cannot find -lvcruntime140
collect2.exe: error: ld returned 1 exit status
Flags ['-msse2', '-mfpmath=sse'] omitted because of compile or link error
C:\MinGW\bin\gcc.exe -mdll -O -Wall -IC:\Anaconda3\include -IC:\Anaconda3\include -c test.c -o test.o -fopenmp
test.c:1:17: fatal error: omp.h: No such file or directory
 #include <omp.h>
                 ^
compilation terminated.
Flags ['-fopenmp', '-fopenmp'] omitted because of compile or link error
skipping 'dipy\reconst\peak_direction_getter.c' Cython extension (up-to-date)
building 'dipy.reconst.peak_direction_getter' extension
C:\MinGW\bin\gcc.exe -mdll -O -Wall -Isrc -IC:\Anaconda3\lib\site-packages\numpy\core\include -Ibuild -IC:\Anaconda3\include -IC:\Anaconda3\include -c dipy\reconst\peak_direction_getter.c -o build\temp.win-amd64-3.5\Release\dipy\reconst\peak_direction_getter.o
In file included from C:\Anaconda3\include/Python.h:65:0,
                 from dipy\reconst\peak_direction_getter.c:4:
C:\Anaconda3\include/pytime.h:113:5: warning: 'struct timeval' declared inside parameter list
     _PyTime_round_t round);
     ^
C:\Anaconda3\include/pytime.h:113:5: warning: its scope is only this definition or declaration, which is probably not what you want
C:\Anaconda3\include/pytime.h:118:5: warning: 'struct timeval' declared inside parameter list
     _PyTime_round_t round);
     ^
In file included from C:\Anaconda3\lib\site-packages\numpy\core\include/numpy/ndarraytypes.h:1781:0,
                 from C:\Anaconda3\lib\site-packages\numpy\core\include/numpy/ndarrayobject.h:18,
                 from C:\Anaconda3\lib\site-packages\numpy\core\include/numpy/arrayobject.h:4,
                 from dipy\reconst\peak_direction_getter.c:242:
C:\Anaconda3\lib\site-packages\numpy\core\include/numpy/npy_1_7_deprecated_api.h:12:9: note: #pragma message: C:\Anaconda3\lib\site-packages\numpy\core\include/numpy/npy_1_7_deprecated_api.h(12) : Warning Msg: Using deprecated NumPy API, disable it by #defining NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION
 #pragma message(_WARN___LOC__"Using deprecated NumPy API, disable it by " \
         ^
In file included from C:\Anaconda3\lib\site-packages\numpy\core\include/numpy/ndarrayobject.h:27:0,
                 from C:\Anaconda3\lib\site-packages\numpy\core\include/numpy/arrayobject.h:4,
                 from dipy\reconst\peak_direction_getter.c:242:
C:\Anaconda3\lib\site-packages\numpy\core\include/numpy/__multiarray_api.h:1634:1: warning: '_import_array' defined but not used [-Wunused-function]
 _import_array(void)
 ^
dipy\reconst\peak_direction_getter.c: In function '__pyx_pw_4dipy_7reconst_21peak_direction_getter_30PeaksAndMetricsDirectionGetter_5initial_direction':
dipy\reconst\peak_direction_getter.c:21688:13: warning: '__pyx_v_numpeaks' may be used uninitialized in this function [-Wmaybe-uninitialized]
             return PyInt_FromLong((long) value);
             ^
dipy\reconst\peak_direction_getter.c:2565:12: note: '__pyx_v_numpeaks' was declared here
   npy_intp __pyx_v_numpeaks;
            ^
writing build\temp.win-amd64-3.5\Release\dipy\reconst\peak_direction_getter.cp35-win_amd64.def
C:\MinGW\bin\gcc.exe -shared -s build\temp.win-amd64-3.5\Release\dipy\reconst\peak_direction_getter.o build\temp.win-amd64-3.5\Release\dipy\reconst\peak_direction_getter.cp35-win_amd64.def -LC:\Anaconda3\libs -LC:\Anaconda3\PCbuild\amd64 -lpython35 -lvcruntime140 -o C:\SDKs\dipy\dipy-head\dipy\dipy\reconst\peak_direction_getter.cp35-win_amd64.pyd
c:/mingw/bin/../lib/gcc/mingw32/4.9.3/../../../../mingw32/bin/ld.exe: C:\Anaconda3\libs/python35.lib(python35.dll): Recognised but unhandled machine type (0x8664) in Import Library Format archive
C:\Anaconda3\libs/python35.lib: error adding symbols: File format not recognized
collect2.exe: error: ld returned 1 exit status
error: command 'C:\\MinGW\\bin\\gcc.exe' failed with exit status 1

From garyfallidis at gmail.com  Mon May 30 18:16:59 2016
From: garyfallidis at gmail.com (Eleftherios Garyfallidis)
Date: Mon, 30 May 2016 22:16:59 +0000
Subject: [Neuroimaging] Fwd: [dipy] Issues trying to install dipy
In-Reply-To: <CAAKhGh800SY0ooqJe9_X1BVnOzbVqjGVtaYOHU-zHGLS1n+1jQ@mail.gmail.com>
References: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>
 <CAAKhGh9-emozuiqQv2KpY8hghks-1b-bid1PQufJ7YUwDffoDQ@mail.gmail.com>
 <CAH6Pt5rVree2gyERp3DSXVmC_dS4j4gQ66YJpouO62W4JpvX=g@mail.gmail.com>
 <CA+X4w0zbG_+5FQx8ALCvij54y+p_4WyfpwfD3LyigWoxAqH66A@mail.gmail.com>
 <CAAKhGh_7yUqHe7EVVmpNvNSbZuv1029CfjvK8Vb3Od57FRssEg@mail.gmail.com>
 <CA+X4w0z=1qjsbLcknNSjMZOkAjrSxhE=8B23E4+_YJPjcJTVrw@mail.gmail.com>
 <CAAKhGh800SY0ooqJe9_X1BVnOzbVqjGVtaYOHU-zHGLS1n+1jQ@mail.gmail.com>
Message-ID: <CAN6JmWOueUqqsiBAHUm9tgyimNxdDUPgJsA7d2aU5d5WxFMzJg@mail.gmail.com>

Hi Jon,

The error is not related to SSE or to OMP. Those are ommitted and then the
compilation continues properly. The problem appears later. Here is the
message

C:\Anaconda3\libs/python35.lib: error adding symbols: File format not recognized
collect2.exe: error: ld returned 1 exit status

Have you contacted the Anaconda developers? This looks like a problem on
their side.

Let us know what they said to you. Otherwise I wonder if this is a specific
problem with Python 3 or if it affects also Python 2. You may want to try
that too. The problem does look more likely to be related to the compiler
used.

Am I correct to say that the only thing that you did was to install
Anaconda and then pip install dipy? Did you have other compilers already
installed in your system?

Best regards,
Eleftherios

On Mon, May 30, 2016 at 5:51 PM Jon Haitz Legarreta <
jhlegarreta at vicomtech.org> wrote:

> Dear Ariel,
> thanks for your suggestion.
>
> The patches in the link seem to help a little bit, but the process seems
> still to be unsuccessful: the MinGW gcc complains with the message:
> gcc: error: /arch:SSE2: No such file or directory
>
> Attached is the new log.
>
> Again, googling was not of much help. I got bits and parts of related
> errors, but have no clear picture of the issue.
>
> Any suggestion is appreciated.
>
> JON HAITZ
>
>
>
>
> On 24 May 2016 at 01:08, Ariel Rokem <arokem at gmail.com> wrote:
>
>>
>>
>> On Mon, May 23, 2016 at 3:19 PM, Jon Haitz Legarreta <
>> jhlegarreta at vicomtech.org> wrote:
>>
>>> Hi there,
>>> thank you Matthew and Ariel.
>>>
>>> The link pointed by Ariel does not seem to be a solution; after having
>>> installed MinGW, as suggested in the link and although I'm aware it might
>>> be unnecessary, the Anaconda3 powershell still yields a similar error, now
>>> pointing to MSVC (which I do not have on my system):
>>>
>>>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 157, in
>>> __init__
>>>     self.dll_libraries = get_msvcr()
>>>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 86, in
>>> get_msvcr
>>>     raise ValueError("Unknown MS Compiler version %s " % msc_ver)
>>> ValueError: Unknown MS Compiler version 1900
>>>
>>> Looks like maybe you ran into this corner case?
>>
>> http://stackoverflow.com/a/34427014/3532933
>>
>>
>>
>>
>>
>>> I'll try to investigate further, and will let you know.
>>>
>>> Kind regards,
>>> JON HAITZ
>>>
>>>
>>>
>>>
>>>
>>>
>>> On 21 May 2016 at 16:55, Ariel Rokem <arokem at gmail.com> wrote:
>>>
>>>> Hi Jon and Matthew,
>>>>
>>>>
>>>> On Sat, May 21, 2016 at 7:30 AM, Matthew Brett <matthew.brett at gmail.com
>>>> > wrote:
>>>>
>>>>> Hi,
>>>>>
>>>>> On Sat, May 21, 2016 at 9:18 AM, Jon Haitz Legarreta
>>>>> <jhlegarreta at vicomtech.org> wrote:
>>>>> > Hi there,
>>>>> > has anybody experienced the issue below?
>>>>> >
>>>>> > Thanks,
>>>>> > JON HAITZ
>>>>> >
>>>>> >
>>>>> >
>>>>> >
>>>>> > ---------- Forwarded message ----------
>>>>> > From: Jon Haitz Legarreta <jhlegarreta at vicomtech.org>
>>>>> > Date: 18 May 2016 at 19:10
>>>>> > Subject: [dipy] Issues trying to install dipy
>>>>> > To: neuroimaging at python.org
>>>>> >
>>>>> >
>>>>> > Hi there,
>>>>> > I'm a newbie to dipy.
>>>>> >
>>>>> > I was trying to follow the instructions in [1] to have dipy
>>>>> installed from
>>>>> > the source code, so that I could execute the dipy examples.
>>>>> >
>>>>> > I'm using Windows 10 and Anaconda 3.
>>>>> >
>>>>> > When trying to execute
>>>>> > python setup.py develop
>>>>> >
>>>>> > the Anaconda prompt yields an error that says in the end:
>>>>> > File: "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126,
>>>>> __init__
>>>>> > if self.ld_version >= "2.10.90":
>>>>> > TypeError: unorderable types: NoneType() >= str()
>>>>> >
>>>>> > I've been googling for a solution without success.
>>>>> >
>>>>> > I don't know whether this looks like Anaconda3 is trying to use
>>>>> cygwin
>>>>> > instead of mingw32, and whether that is the root cause.
>>>>> >
>>>>> > In either case, does anyone know how to solve the issue?
>>>>> >
>>>>> > Attached is the trace (it's short) of the error if this is of any
>>>>> help.
>>>>> >
>>>>> >
>>>>> > Thank you,
>>>>> > JON HAITZ
>>>>> >
>>>>> >
>>>>> > [1] http://nipy.org/dipy/installation.html#install-source-nix
>>>>>
>>>>> I'm sorry, I'm afraid I don't personally use Anaconda, so I have no
>>>>> experience of fixing compilation errors on Anaconda.   Ariel - have
>>>>> you come across this?
>>>>>
>>>>
>>>> And I don't personally use Windows...
>>>>
>>>> Might this be helpful:
>>>>
>>>>
>>>> http://stackoverflow.com/questions/24683305/python-cant-install-packages-typeerror-unorderable-types-nonetype-str
>>>>
>>>> It seems like it could be related, though it's all Greek to me.
>>>>
>>>> Cheers,
>>>>
>>>> Ariel
>>>>
>>>>
>>>>> You could also try on the anaconda support channels (issues, mailing
>>>>> list) - it may well be a general problem rather than one specific to
>>>>> dipy,
>>>>>
>>>>> Best,
>>>>>
>>>>> Matthew
>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>>
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160530/a84d543c/attachment.html>

From matthew.brett at gmail.com  Mon May 30 21:45:27 2016
From: matthew.brett at gmail.com (Matthew Brett)
Date: Mon, 30 May 2016 18:45:27 -0700
Subject: [Neuroimaging] NiBabel - Streamlines API - TRK header extension
In-Reply-To: <5746E704.5040100@gmail.com>
References: <5746E704.5040100@gmail.com>
Message-ID: <CAH6Pt5oYRexN+y8aMjvs-+RQLTnrFV7OZaUABBPLq71qaFXF1Q@mail.gmail.com>

Hi,

On Thu, May 26, 2016 at 5:07 AM, Marc <marc.cote.19 at gmail.com> wrote:

> Hi everyone,
>
> I'm currently working on an API for streamlines in NiBabel and I would
> like the feedback of the Nipy community about an unofficial extension for
> the TRK header.
>
> Right now, the TrackVis's file format for streamlines supports conserving
> additional information for each point and/or streamline. The format refers
> to that additional information as "scalars" and "properties" respectively
> (ref: http://www.trackvis.org/docs/?subsect=fileformat). There is a field
> in the header to keep a tag name for the first 20 scalars and the first 20
> properties even though the format supports having up to 2^16 scalars and
> 2^16 properties.
>
> First, 20 is not a lot when you think of all the metrics one could want to
> keep (e.g. color, FA, curvature, torsion, MD, cluster ID, etc.) in its
> tractogram file. On top of that, some of these metrics consist of more than
> one value, for instance the color is in fact composed of three values (i.e.
> RGB). I find it a bit wasteful to use up multiple tag names only because it
> is composed of multiple values.
>
> What I proposed is to encode that number of values in the tag name at save
> time and take that encoded number into account when loading a TRK file. I
> have two ways in mind and I would like your opinion.
>
> *1st approach*
> A tag name can only have 20 characters/bytes. I would use the last two
> bytes to encode the number of values associated to a given tag name. The
> first byte would always be \x00 (EOL) so that TrackVis doesn't try to
> interpret the last byte which will be set to the number of values (since it
> is a uint8 the range is [0, 255]). The upside of this approach is that
> TrackVis is not aware (read doesn't crash) thanks to the \x00 "hack". The
> downside is that a TrackVis's user will be less aware of what's going on
> behind the scene but a NiBabel's user won't see a difference since we
> revert the process when loading the TRK file.
>

Just to check - how can you be sure when you find \x00 \x04 at the end of
the trackvis tag name, that this is really telling you this is a nibabel
hacked name with 4 values, rather than random cruft in the file?

Cheers,

Matthew
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160530/fd80777f/attachment.html>

From marc.cote.19 at gmail.com  Tue May 31 14:32:47 2016
From: marc.cote.19 at gmail.com (=?UTF-8?B?TWFyYy1BbGV4YW5kcmUgQ8O0dMOp?=)
Date: Tue, 31 May 2016 14:32:47 -0400
Subject: [Neuroimaging] NiBabel - Streamlines API - TRK header extension
In-Reply-To: <CAH6Pt5oYRexN+y8aMjvs-+RQLTnrFV7OZaUABBPLq71qaFXF1Q@mail.gmail.com>
References: <5746E704.5040100@gmail.com>
 <CAH6Pt5oYRexN+y8aMjvs-+RQLTnrFV7OZaUABBPLq71qaFXF1Q@mail.gmail.com>
Message-ID: <574DD8CF.4030806@gmail.com>

You are right there is no guarantee. I rely on the common sense of 
initializing memory to \x00. Especially, when that memory corresponds to 
a documented header field.

A more robust approach could be to replace \x00 by a checksum of the tag 
name. However, I don't think it is worth doing it right now.

Marc

On 16-05-30 09:45 PM, Matthew Brett wrote:
> Hi,
>
> On Thu, May 26, 2016 at 5:07 AM, Marc <marc.cote.19 at gmail.com 
> <mailto:marc.cote.19 at gmail.com>> wrote:
>
>     Hi everyone,
>
>     I'm currently working on an API for streamlines in NiBabel and I
>     would like the feedback of the Nipy community about an unofficial
>     extension for the TRK header.
>
>     Right now, the TrackVis's file format for streamlines supports
>     conserving additional information for each point and/or
>     streamline. The format refers to that additional information as
>     "scalars" and "properties" respectively (ref:
>     http://www.trackvis.org/docs/?subsect=fileformat). There is a
>     field in the header to keep a tag name for the first 20 scalars
>     and the first 20 properties even though the format supports having
>     up to 2^16 scalars and 2^16 properties.
>
>     First, 20 is not a lot when you think of all the metrics one could
>     want to keep (e.g. color, FA, curvature, torsion, MD, cluster ID,
>     etc.) in its tractogram file. On top of that, some of these
>     metrics consist of more than one value, for instance the color is
>     in fact composed of three values (i.e. RGB). I find it a bit
>     wasteful to use up multiple tag names only because it is composed
>     of multiple values.
>
>     What I proposed is to encode that number of values in the tag name
>     at save time and take that encoded number into account when
>     loading a TRK file. I have two ways in mind and I would like your
>     opinion.
>
>     *1st approach*
>     A tag name can only have 20 characters/bytes. I would use the last
>     two bytes to encode the number of values associated to a given tag
>     name. The first byte would always be \x00 (EOL) so that TrackVis
>     doesn't try to interpret the last byte which will be set to the
>     number of values (since it is a uint8 the range is [0, 255]). The
>     upside of this approach is that TrackVis is not aware (read
>     doesn't crash) thanks to the \x00 "hack". The downside is that a
>     TrackVis's user will be less aware of what's going on behind the
>     scene but a NiBabel's user won't see a difference since we revert
>     the process when loading the TRK file.
>
>
> Just to check - how can you be sure when you find \x00 \x04 at the end 
> of the trackvis tag name, that this is really telling you this is a 
> nibabel hacked name with 4 values, rather than random cruft in the file?
>
> Cheers,
>
> Matthew
>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160531/22d5dba3/attachment.html>

From matthew.brett at gmail.com  Tue May 31 14:55:01 2016
From: matthew.brett at gmail.com (Matthew Brett)
Date: Tue, 31 May 2016 11:55:01 -0700
Subject: [Neuroimaging] NiBabel - Streamlines API - TRK header extension
In-Reply-To: <574DD8CF.4030806@gmail.com>
References: <5746E704.5040100@gmail.com>
 <CAH6Pt5oYRexN+y8aMjvs-+RQLTnrFV7OZaUABBPLq71qaFXF1Q@mail.gmail.com>
 <574DD8CF.4030806@gmail.com>
Message-ID: <CAH6Pt5peEK3ptkAXz6shGm=ZKNyQQPHFELrn1n7GGFTBFUY6Cw@mail.gmail.com>

On Tue, May 31, 2016 at 11:32 AM, Marc-Alexandre C?t?
<marc.cote.19 at gmail.com> wrote:
> You are right there is no guarantee. I rely on the common sense of
> initializing memory to \x00. Especially, when that memory corresponds to a
> documented header field.
>
> A more robust approach could be to replace \x00 by a checksum of the tag
> name. However, I don't think it is worth doing it right now.

Well - I guess the checksum could end up being a valid ascii char.

Other options are:

* add a magic value after the \x0 and before the number of properties;
* use ascii for the number N, as in `name\ x0 8 \x0 \x0 ... `.  This
has the advantage that it's a little easier for the person reading
this file to guess what's going on, uses no extra bytes when N<10, and
a check for [ascii name] then [\x0] then [ascii number] then [\x0 or
end of field] would be reasonably reliable as check against random
bytes in the fields.

Matthew

From jhlegarreta at vicomtech.org  Tue May 31 14:58:08 2016
From: jhlegarreta at vicomtech.org (Jon Haitz Legarreta)
Date: Tue, 31 May 2016 20:58:08 +0200
Subject: [Neuroimaging] Fwd: [dipy] Issues trying to install dipy
In-Reply-To: <CAN6JmWOueUqqsiBAHUm9tgyimNxdDUPgJsA7d2aU5d5WxFMzJg@mail.gmail.com>
References: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>
 <CAAKhGh9-emozuiqQv2KpY8hghks-1b-bid1PQufJ7YUwDffoDQ@mail.gmail.com>
 <CAH6Pt5rVree2gyERp3DSXVmC_dS4j4gQ66YJpouO62W4JpvX=g@mail.gmail.com>
 <CA+X4w0zbG_+5FQx8ALCvij54y+p_4WyfpwfD3LyigWoxAqH66A@mail.gmail.com>
 <CAAKhGh_7yUqHe7EVVmpNvNSbZuv1029CfjvK8Vb3Od57FRssEg@mail.gmail.com>
 <CA+X4w0z=1qjsbLcknNSjMZOkAjrSxhE=8B23E4+_YJPjcJTVrw@mail.gmail.com>
 <CAAKhGh800SY0ooqJe9_X1BVnOzbVqjGVtaYOHU-zHGLS1n+1jQ@mail.gmail.com>
 <CAN6JmWOueUqqsiBAHUm9tgyimNxdDUPgJsA7d2aU5d5WxFMzJg@mail.gmail.com>
Message-ID: <CAAKhGh9x5Jhp78FQKaExxYSCAOBh3qxzVWfy+nfPskf0dA0Lnw@mail.gmail.com>

Dear Eleftherios,
thanks for the follow-up.

No, I did not contact the Anaconda developers, since I was a little bit
lost in the messages/source problem. But if your guess is that it is more
likely an Anaconda problem, I will contact them and let you know.

On the other hand, you are right; I installed Anaconda, then tried to set
up dipy for development.

Also I have a virtual environment where I downloaded the dipy dependencies.
I do not know if the latter step can be avoided or else, whether one can be
substituted by the other: i.e. and whether when one installs dipy for
development from sources, dipy scripts take care of putting in place the
necessary packages. I just thought that creating a virtual env with just
the dipy dependencies would be cleaner for development (i.e. avoid clashes
with other packages or repos, etc.)

Since one of the links posted suggested it, I installed MinGW, but had no
other compiler installer on my machine.

Sincerely,
JON HAITZ



On 31 May 2016 at 00:16, Eleftherios Garyfallidis <garyfallidis at gmail.com>
wrote:

> Hi Jon,
>
> The error is not related to SSE or to OMP. Those are ommitted and then the
> compilation continues properly. The problem appears later. Here is the
> message
>
> C:\Anaconda3\libs/python35.lib: error adding symbols: File format not recognized
> collect2.exe: error: ld returned 1 exit status
>
> Have you contacted the Anaconda developers? This looks like a problem on
> their side.
>
> Let us know what they said to you. Otherwise I wonder if this is a
> specific problem with Python 3 or if it affects also Python 2. You may want
> to try that too. The problem does look more likely to be related to the
> compiler used.
>
> Am I correct to say that the only thing that you did was to install
> Anaconda and then pip install dipy? Did you have other compilers already
> installed in your system?
>
> Best regards,
> Eleftherios
>
> On Mon, May 30, 2016 at 5:51 PM Jon Haitz Legarreta <
> jhlegarreta at vicomtech.org> wrote:
>
>> Dear Ariel,
>> thanks for your suggestion.
>>
>> The patches in the link seem to help a little bit, but the process seems
>> still to be unsuccessful: the MinGW gcc complains with the message:
>> gcc: error: /arch:SSE2: No such file or directory
>>
>> Attached is the new log.
>>
>> Again, googling was not of much help. I got bits and parts of related
>> errors, but have no clear picture of the issue.
>>
>> Any suggestion is appreciated.
>>
>> JON HAITZ
>>
>>
>>
>>
>> On 24 May 2016 at 01:08, Ariel Rokem <arokem at gmail.com> wrote:
>>
>>>
>>>
>>> On Mon, May 23, 2016 at 3:19 PM, Jon Haitz Legarreta <
>>> jhlegarreta at vicomtech.org> wrote:
>>>
>>>> Hi there,
>>>> thank you Matthew and Ariel.
>>>>
>>>> The link pointed by Ariel does not seem to be a solution; after having
>>>> installed MinGW, as suggested in the link and although I'm aware it might
>>>> be unnecessary, the Anaconda3 powershell still yields a similar error, now
>>>> pointing to MSVC (which I do not have on my system):
>>>>
>>>>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 157, in
>>>> __init__
>>>>     self.dll_libraries = get_msvcr()
>>>>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 86, in
>>>> get_msvcr
>>>>     raise ValueError("Unknown MS Compiler version %s " % msc_ver)
>>>> ValueError: Unknown MS Compiler version 1900
>>>>
>>>> Looks like maybe you ran into this corner case?
>>>
>>> http://stackoverflow.com/a/34427014/3532933
>>>
>>>
>>>
>>>
>>>
>>>> I'll try to investigate further, and will let you know.
>>>>
>>>> Kind regards,
>>>> JON HAITZ
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On 21 May 2016 at 16:55, Ariel Rokem <arokem at gmail.com> wrote:
>>>>
>>>>> Hi Jon and Matthew,
>>>>>
>>>>>
>>>>> On Sat, May 21, 2016 at 7:30 AM, Matthew Brett <
>>>>> matthew.brett at gmail.com> wrote:
>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> On Sat, May 21, 2016 at 9:18 AM, Jon Haitz Legarreta
>>>>>> <jhlegarreta at vicomtech.org> wrote:
>>>>>> > Hi there,
>>>>>> > has anybody experienced the issue below?
>>>>>> >
>>>>>> > Thanks,
>>>>>> > JON HAITZ
>>>>>> >
>>>>>> >
>>>>>> >
>>>>>> >
>>>>>> > ---------- Forwarded message ----------
>>>>>> > From: Jon Haitz Legarreta <jhlegarreta at vicomtech.org>
>>>>>> > Date: 18 May 2016 at 19:10
>>>>>> > Subject: [dipy] Issues trying to install dipy
>>>>>> > To: neuroimaging at python.org
>>>>>> >
>>>>>> >
>>>>>> > Hi there,
>>>>>> > I'm a newbie to dipy.
>>>>>> >
>>>>>> > I was trying to follow the instructions in [1] to have dipy
>>>>>> installed from
>>>>>> > the source code, so that I could execute the dipy examples.
>>>>>> >
>>>>>> > I'm using Windows 10 and Anaconda 3.
>>>>>> >
>>>>>> > When trying to execute
>>>>>> > python setup.py develop
>>>>>> >
>>>>>> > the Anaconda prompt yields an error that says in the end:
>>>>>> > File: "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126,
>>>>>> __init__
>>>>>> > if self.ld_version >= "2.10.90":
>>>>>> > TypeError: unorderable types: NoneType() >= str()
>>>>>> >
>>>>>> > I've been googling for a solution without success.
>>>>>> >
>>>>>> > I don't know whether this looks like Anaconda3 is trying to use
>>>>>> cygwin
>>>>>> > instead of mingw32, and whether that is the root cause.
>>>>>> >
>>>>>> > In either case, does anyone know how to solve the issue?
>>>>>> >
>>>>>> > Attached is the trace (it's short) of the error if this is of any
>>>>>> help.
>>>>>> >
>>>>>> >
>>>>>> > Thank you,
>>>>>> > JON HAITZ
>>>>>> >
>>>>>> >
>>>>>> > [1] http://nipy.org/dipy/installation.html#install-source-nix
>>>>>>
>>>>>> I'm sorry, I'm afraid I don't personally use Anaconda, so I have no
>>>>>> experience of fixing compilation errors on Anaconda.   Ariel - have
>>>>>> you come across this?
>>>>>>
>>>>>
>>>>> And I don't personally use Windows...
>>>>>
>>>>> Might this be helpful:
>>>>>
>>>>>
>>>>> http://stackoverflow.com/questions/24683305/python-cant-install-packages-typeerror-unorderable-types-nonetype-str
>>>>>
>>>>> It seems like it could be related, though it's all Greek to me.
>>>>>
>>>>> Cheers,
>>>>>
>>>>> Ariel
>>>>>
>>>>>
>>>>>> You could also try on the anaconda support channels (issues, mailing
>>>>>> list) - it may well be a general problem rather than one specific to
>>>>>> dipy,
>>>>>>
>>>>>> Best,
>>>>>>
>>>>>> Matthew
>>>>>> _______________________________________________
>>>>>> Neuroimaging mailing list
>>>>>> Neuroimaging at python.org
>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>>
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160531/5a6a00d0/attachment.html>

From grlee77 at gmail.com  Tue May 31 16:01:52 2016
From: grlee77 at gmail.com (Gregory Lee)
Date: Tue, 31 May 2016 16:01:52 -0400
Subject: [Neuroimaging] Fwd: [dipy] Issues trying to install dipy
In-Reply-To: <CAAKhGh9x5Jhp78FQKaExxYSCAOBh3qxzVWfy+nfPskf0dA0Lnw@mail.gmail.com>
References: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>
 <CAAKhGh9-emozuiqQv2KpY8hghks-1b-bid1PQufJ7YUwDffoDQ@mail.gmail.com>
 <CAH6Pt5rVree2gyERp3DSXVmC_dS4j4gQ66YJpouO62W4JpvX=g@mail.gmail.com>
 <CA+X4w0zbG_+5FQx8ALCvij54y+p_4WyfpwfD3LyigWoxAqH66A@mail.gmail.com>
 <CAAKhGh_7yUqHe7EVVmpNvNSbZuv1029CfjvK8Vb3Od57FRssEg@mail.gmail.com>
 <CA+X4w0z=1qjsbLcknNSjMZOkAjrSxhE=8B23E4+_YJPjcJTVrw@mail.gmail.com>
 <CAAKhGh800SY0ooqJe9_X1BVnOzbVqjGVtaYOHU-zHGLS1n+1jQ@mail.gmail.com>
 <CAN6JmWOueUqqsiBAHUm9tgyimNxdDUPgJsA7d2aU5d5WxFMzJg@mail.gmail.com>
 <CAAKhGh9x5Jhp78FQKaExxYSCAOBh3qxzVWfy+nfPskf0dA0Lnw@mail.gmail.com>
Message-ID: <CAJR3sXd3ESkxq44qgaiKRPr_ROwQpT-sZTLrGcb7OGxjGOVLdg@mail.gmail.com>

Hi Jon,

Do you need to build from the lastest source or are you just trying to
install a working version to run examples?  If the later, it may be worth
trying the pre-built conda packages Ariel recently created at conda-forge.
To install:

conda install -c conda-forge dipy

- Greg

On Tue, May 31, 2016 at 2:58 PM, Jon Haitz Legarreta <
jhlegarreta at vicomtech.org> wrote:

> Dear Eleftherios,
> thanks for the follow-up.
>
> No, I did not contact the Anaconda developers, since I was a little bit
> lost in the messages/source problem. But if your guess is that it is more
> likely an Anaconda problem, I will contact them and let you know.
>
> On the other hand, you are right; I installed Anaconda, then tried to set
> up dipy for development.
>
> Also I have a virtual environment where I downloaded the dipy
> dependencies. I do not know if the latter step can be avoided or else,
> whether one can be substituted by the other: i.e. and whether when one
> installs dipy for development from sources, dipy scripts take care of
> putting in place the necessary packages. I just thought that creating a
> virtual env with just the dipy dependencies would be cleaner for
> development (i.e. avoid clashes with other packages or repos, etc.)
>
> Since one of the links posted suggested it, I installed MinGW, but had no
> other compiler installer on my machine.
>
> Sincerely,
> JON HAITZ
>
>
>
>
> On 31 May 2016 at 00:16, Eleftherios Garyfallidis <garyfallidis at gmail.com>
> wrote:
>
>> Hi Jon,
>>
>> The error is not related to SSE or to OMP. Those are ommitted and then
>> the compilation continues properly. The problem appears later. Here is the
>> message
>>
>> C:\Anaconda3\libs/python35.lib: error adding symbols: File format not recognized
>> collect2.exe: error: ld returned 1 exit status
>>
>> Have you contacted the Anaconda developers? This looks like a problem on
>> their side.
>>
>> Let us know what they said to you. Otherwise I wonder if this is a
>> specific problem with Python 3 or if it affects also Python 2. You may want
>> to try that too. The problem does look more likely to be related to the
>> compiler used.
>>
>> Am I correct to say that the only thing that you did was to install
>> Anaconda and then pip install dipy? Did you have other compilers already
>> installed in your system?
>>
>> Best regards,
>> Eleftherios
>>
>> On Mon, May 30, 2016 at 5:51 PM Jon Haitz Legarreta <
>> jhlegarreta at vicomtech.org> wrote:
>>
>>> Dear Ariel,
>>> thanks for your suggestion.
>>>
>>> The patches in the link seem to help a little bit, but the process seems
>>> still to be unsuccessful: the MinGW gcc complains with the message:
>>> gcc: error: /arch:SSE2: No such file or directory
>>>
>>> Attached is the new log.
>>>
>>> Again, googling was not of much help. I got bits and parts of related
>>> errors, but have no clear picture of the issue.
>>>
>>> Any suggestion is appreciated.
>>>
>>> JON HAITZ
>>>
>>>
>>>
>>>
>>> On 24 May 2016 at 01:08, Ariel Rokem <arokem at gmail.com> wrote:
>>>
>>>>
>>>>
>>>> On Mon, May 23, 2016 at 3:19 PM, Jon Haitz Legarreta <
>>>> jhlegarreta at vicomtech.org> wrote:
>>>>
>>>>> Hi there,
>>>>> thank you Matthew and Ariel.
>>>>>
>>>>> The link pointed by Ariel does not seem to be a solution; after having
>>>>> installed MinGW, as suggested in the link and although I'm aware it might
>>>>> be unnecessary, the Anaconda3 powershell still yields a similar error, now
>>>>> pointing to MSVC (which I do not have on my system):
>>>>>
>>>>>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 157, in
>>>>> __init__
>>>>>     self.dll_libraries = get_msvcr()
>>>>>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 86, in
>>>>> get_msvcr
>>>>>     raise ValueError("Unknown MS Compiler version %s " % msc_ver)
>>>>> ValueError: Unknown MS Compiler version 1900
>>>>>
>>>>> Looks like maybe you ran into this corner case?
>>>>
>>>> http://stackoverflow.com/a/34427014/3532933
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>> I'll try to investigate further, and will let you know.
>>>>>
>>>>> Kind regards,
>>>>> JON HAITZ
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 21 May 2016 at 16:55, Ariel Rokem <arokem at gmail.com> wrote:
>>>>>
>>>>>> Hi Jon and Matthew,
>>>>>>
>>>>>>
>>>>>> On Sat, May 21, 2016 at 7:30 AM, Matthew Brett <
>>>>>> matthew.brett at gmail.com> wrote:
>>>>>>
>>>>>>> Hi,
>>>>>>>
>>>>>>> On Sat, May 21, 2016 at 9:18 AM, Jon Haitz Legarreta
>>>>>>> <jhlegarreta at vicomtech.org> wrote:
>>>>>>> > Hi there,
>>>>>>> > has anybody experienced the issue below?
>>>>>>> >
>>>>>>> > Thanks,
>>>>>>> > JON HAITZ
>>>>>>> >
>>>>>>> >
>>>>>>> >
>>>>>>> >
>>>>>>> > ---------- Forwarded message ----------
>>>>>>> > From: Jon Haitz Legarreta <jhlegarreta at vicomtech.org>
>>>>>>> > Date: 18 May 2016 at 19:10
>>>>>>> > Subject: [dipy] Issues trying to install dipy
>>>>>>> > To: neuroimaging at python.org
>>>>>>> >
>>>>>>> >
>>>>>>> > Hi there,
>>>>>>> > I'm a newbie to dipy.
>>>>>>> >
>>>>>>> > I was trying to follow the instructions in [1] to have dipy
>>>>>>> installed from
>>>>>>> > the source code, so that I could execute the dipy examples.
>>>>>>> >
>>>>>>> > I'm using Windows 10 and Anaconda 3.
>>>>>>> >
>>>>>>> > When trying to execute
>>>>>>> > python setup.py develop
>>>>>>> >
>>>>>>> > the Anaconda prompt yields an error that says in the end:
>>>>>>> > File: "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126,
>>>>>>> __init__
>>>>>>> > if self.ld_version >= "2.10.90":
>>>>>>> > TypeError: unorderable types: NoneType() >= str()
>>>>>>> >
>>>>>>> > I've been googling for a solution without success.
>>>>>>> >
>>>>>>> > I don't know whether this looks like Anaconda3 is trying to use
>>>>>>> cygwin
>>>>>>> > instead of mingw32, and whether that is the root cause.
>>>>>>> >
>>>>>>> > In either case, does anyone know how to solve the issue?
>>>>>>> >
>>>>>>> > Attached is the trace (it's short) of the error if this is of any
>>>>>>> help.
>>>>>>> >
>>>>>>> >
>>>>>>> > Thank you,
>>>>>>> > JON HAITZ
>>>>>>> >
>>>>>>> >
>>>>>>> > [1] http://nipy.org/dipy/installation.html#install-source-nix
>>>>>>>
>>>>>>> I'm sorry, I'm afraid I don't personally use Anaconda, so I have no
>>>>>>> experience of fixing compilation errors on Anaconda.   Ariel - have
>>>>>>> you come across this?
>>>>>>>
>>>>>>
>>>>>> And I don't personally use Windows...
>>>>>>
>>>>>> Might this be helpful:
>>>>>>
>>>>>>
>>>>>> http://stackoverflow.com/questions/24683305/python-cant-install-packages-typeerror-unorderable-types-nonetype-str
>>>>>>
>>>>>> It seems like it could be related, though it's all Greek to me.
>>>>>>
>>>>>> Cheers,
>>>>>>
>>>>>> Ariel
>>>>>>
>>>>>>
>>>>>>> You could also try on the anaconda support channels (issues, mailing
>>>>>>> list) - it may well be a general problem rather than one specific to
>>>>>>> dipy,
>>>>>>>
>>>>>>> Best,
>>>>>>>
>>>>>>> Matthew
>>>>>>> _______________________________________________
>>>>>>> Neuroimaging mailing list
>>>>>>> Neuroimaging at python.org
>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Neuroimaging mailing list
>>>>>> Neuroimaging at python.org
>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160531/d12e4f3a/attachment.html>

From jhlegarreta at vicomtech.org  Tue May 31 17:33:49 2016
From: jhlegarreta at vicomtech.org (Jon Haitz Legarreta)
Date: Tue, 31 May 2016 23:33:49 +0200
Subject: [Neuroimaging] Fwd: [dipy] Issues trying to install dipy
In-Reply-To: <CAJR3sXd3ESkxq44qgaiKRPr_ROwQpT-sZTLrGcb7OGxjGOVLdg@mail.gmail.com>
References: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>
 <CAAKhGh9-emozuiqQv2KpY8hghks-1b-bid1PQufJ7YUwDffoDQ@mail.gmail.com>
 <CAH6Pt5rVree2gyERp3DSXVmC_dS4j4gQ66YJpouO62W4JpvX=g@mail.gmail.com>
 <CA+X4w0zbG_+5FQx8ALCvij54y+p_4WyfpwfD3LyigWoxAqH66A@mail.gmail.com>
 <CAAKhGh_7yUqHe7EVVmpNvNSbZuv1029CfjvK8Vb3Od57FRssEg@mail.gmail.com>
 <CA+X4w0z=1qjsbLcknNSjMZOkAjrSxhE=8B23E4+_YJPjcJTVrw@mail.gmail.com>
 <CAAKhGh800SY0ooqJe9_X1BVnOzbVqjGVtaYOHU-zHGLS1n+1jQ@mail.gmail.com>
 <CAN6JmWOueUqqsiBAHUm9tgyimNxdDUPgJsA7d2aU5d5WxFMzJg@mail.gmail.com>
 <CAAKhGh9x5Jhp78FQKaExxYSCAOBh3qxzVWfy+nfPskf0dA0Lnw@mail.gmail.com>
 <CAJR3sXd3ESkxq44qgaiKRPr_ROwQpT-sZTLrGcb7OGxjGOVLdg@mail.gmail.com>
Message-ID: <CAAKhGh-y=eMep32OYsrEyr_XhxKD_9p4RVBgK5RAu6CxkRiVag@mail.gmail.com>

Hi Greg,
using the virtual env I'm able to run the examples without major issues.
Indeed, for the virtual env I downloaded the conda-forge dipy package.

But my intent is to build from the latest source.

Thanks,
JON HAITZ




On 31 May 2016 at 22:01, Gregory Lee <grlee77 at gmail.com> wrote:

> Hi Jon,
>
> Do you need to build from the lastest source or are you just trying to
> install a working version to run examples?  If the later, it may be worth
> trying the pre-built conda packages Ariel recently created at conda-forge.
> To install:
>
> conda install -c conda-forge dipy
>
> - Greg
>
> On Tue, May 31, 2016 at 2:58 PM, Jon Haitz Legarreta <
> jhlegarreta at vicomtech.org> wrote:
>
>> Dear Eleftherios,
>> thanks for the follow-up.
>>
>> No, I did not contact the Anaconda developers, since I was a little bit
>> lost in the messages/source problem. But if your guess is that it is more
>> likely an Anaconda problem, I will contact them and let you know.
>>
>> On the other hand, you are right; I installed Anaconda, then tried to set
>> up dipy for development.
>>
>> Also I have a virtual environment where I downloaded the dipy
>> dependencies. I do not know if the latter step can be avoided or else,
>> whether one can be substituted by the other: i.e. and whether when one
>> installs dipy for development from sources, dipy scripts take care of
>> putting in place the necessary packages. I just thought that creating a
>> virtual env with just the dipy dependencies would be cleaner for
>> development (i.e. avoid clashes with other packages or repos, etc.)
>>
>> Since one of the links posted suggested it, I installed MinGW, but had no
>> other compiler installer on my machine.
>>
>> Sincerely,
>> JON HAITZ
>>
>>
>>
>>
>> On 31 May 2016 at 00:16, Eleftherios Garyfallidis <garyfallidis at gmail.com
>> > wrote:
>>
>>> Hi Jon,
>>>
>>> The error is not related to SSE or to OMP. Those are ommitted and then
>>> the compilation continues properly. The problem appears later. Here is the
>>> message
>>>
>>> C:\Anaconda3\libs/python35.lib: error adding symbols: File format not recognized
>>> collect2.exe: error: ld returned 1 exit status
>>>
>>> Have you contacted the Anaconda developers? This looks like a problem on
>>> their side.
>>>
>>> Let us know what they said to you. Otherwise I wonder if this is a
>>> specific problem with Python 3 or if it affects also Python 2. You may want
>>> to try that too. The problem does look more likely to be related to the
>>> compiler used.
>>>
>>> Am I correct to say that the only thing that you did was to install
>>> Anaconda and then pip install dipy? Did you have other compilers already
>>> installed in your system?
>>>
>>> Best regards,
>>> Eleftherios
>>>
>>> On Mon, May 30, 2016 at 5:51 PM Jon Haitz Legarreta <
>>> jhlegarreta at vicomtech.org> wrote:
>>>
>>>> Dear Ariel,
>>>> thanks for your suggestion.
>>>>
>>>> The patches in the link seem to help a little bit, but the process
>>>> seems still to be unsuccessful: the MinGW gcc complains with the message:
>>>> gcc: error: /arch:SSE2: No such file or directory
>>>>
>>>> Attached is the new log.
>>>>
>>>> Again, googling was not of much help. I got bits and parts of related
>>>> errors, but have no clear picture of the issue.
>>>>
>>>> Any suggestion is appreciated.
>>>>
>>>> JON HAITZ
>>>>
>>>>
>>>>
>>>>
>>>> On 24 May 2016 at 01:08, Ariel Rokem <arokem at gmail.com> wrote:
>>>>
>>>>>
>>>>>
>>>>> On Mon, May 23, 2016 at 3:19 PM, Jon Haitz Legarreta <
>>>>> jhlegarreta at vicomtech.org> wrote:
>>>>>
>>>>>> Hi there,
>>>>>> thank you Matthew and Ariel.
>>>>>>
>>>>>> The link pointed by Ariel does not seem to be a solution; after
>>>>>> having installed MinGW, as suggested in the link and although I'm aware it
>>>>>> might be unnecessary, the Anaconda3 powershell still yields a similar
>>>>>> error, now pointing to MSVC (which I do not have on my system):
>>>>>>
>>>>>>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 157, in
>>>>>> __init__
>>>>>>     self.dll_libraries = get_msvcr()
>>>>>>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 86, in
>>>>>> get_msvcr
>>>>>>     raise ValueError("Unknown MS Compiler version %s " % msc_ver)
>>>>>> ValueError: Unknown MS Compiler version 1900
>>>>>>
>>>>>> Looks like maybe you ran into this corner case?
>>>>>
>>>>> http://stackoverflow.com/a/34427014/3532933
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>> I'll try to investigate further, and will let you know.
>>>>>>
>>>>>> Kind regards,
>>>>>> JON HAITZ
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> On 21 May 2016 at 16:55, Ariel Rokem <arokem at gmail.com> wrote:
>>>>>>
>>>>>>> Hi Jon and Matthew,
>>>>>>>
>>>>>>>
>>>>>>> On Sat, May 21, 2016 at 7:30 AM, Matthew Brett <
>>>>>>> matthew.brett at gmail.com> wrote:
>>>>>>>
>>>>>>>> Hi,
>>>>>>>>
>>>>>>>> On Sat, May 21, 2016 at 9:18 AM, Jon Haitz Legarreta
>>>>>>>> <jhlegarreta at vicomtech.org> wrote:
>>>>>>>> > Hi there,
>>>>>>>> > has anybody experienced the issue below?
>>>>>>>> >
>>>>>>>> > Thanks,
>>>>>>>> > JON HAITZ
>>>>>>>> >
>>>>>>>> >
>>>>>>>> >
>>>>>>>> >
>>>>>>>> > ---------- Forwarded message ----------
>>>>>>>> > From: Jon Haitz Legarreta <jhlegarreta at vicomtech.org>
>>>>>>>> > Date: 18 May 2016 at 19:10
>>>>>>>> > Subject: [dipy] Issues trying to install dipy
>>>>>>>> > To: neuroimaging at python.org
>>>>>>>> >
>>>>>>>> >
>>>>>>>> > Hi there,
>>>>>>>> > I'm a newbie to dipy.
>>>>>>>> >
>>>>>>>> > I was trying to follow the instructions in [1] to have dipy
>>>>>>>> installed from
>>>>>>>> > the source code, so that I could execute the dipy examples.
>>>>>>>> >
>>>>>>>> > I'm using Windows 10 and Anaconda 3.
>>>>>>>> >
>>>>>>>> > When trying to execute
>>>>>>>> > python setup.py develop
>>>>>>>> >
>>>>>>>> > the Anaconda prompt yields an error that says in the end:
>>>>>>>> > File: "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126,
>>>>>>>> __init__
>>>>>>>> > if self.ld_version >= "2.10.90":
>>>>>>>> > TypeError: unorderable types: NoneType() >= str()
>>>>>>>> >
>>>>>>>> > I've been googling for a solution without success.
>>>>>>>> >
>>>>>>>> > I don't know whether this looks like Anaconda3 is trying to use
>>>>>>>> cygwin
>>>>>>>> > instead of mingw32, and whether that is the root cause.
>>>>>>>> >
>>>>>>>> > In either case, does anyone know how to solve the issue?
>>>>>>>> >
>>>>>>>> > Attached is the trace (it's short) of the error if this is of any
>>>>>>>> help.
>>>>>>>> >
>>>>>>>> >
>>>>>>>> > Thank you,
>>>>>>>> > JON HAITZ
>>>>>>>> >
>>>>>>>> >
>>>>>>>> > [1] http://nipy.org/dipy/installation.html#install-source-nix
>>>>>>>>
>>>>>>>> I'm sorry, I'm afraid I don't personally use Anaconda, so I have no
>>>>>>>> experience of fixing compilation errors on Anaconda.   Ariel - have
>>>>>>>> you come across this?
>>>>>>>>
>>>>>>>
>>>>>>> And I don't personally use Windows...
>>>>>>>
>>>>>>> Might this be helpful:
>>>>>>>
>>>>>>>
>>>>>>> http://stackoverflow.com/questions/24683305/python-cant-install-packages-typeerror-unorderable-types-nonetype-str
>>>>>>>
>>>>>>> It seems like it could be related, though it's all Greek to me.
>>>>>>>
>>>>>>> Cheers,
>>>>>>>
>>>>>>> Ariel
>>>>>>>
>>>>>>>
>>>>>>>> You could also try on the anaconda support channels (issues, mailing
>>>>>>>> list) - it may well be a general problem rather than one specific to
>>>>>>>> dipy,
>>>>>>>>
>>>>>>>> Best,
>>>>>>>>
>>>>>>>> Matthew
>>>>>>>> _______________________________________________
>>>>>>>> Neuroimaging mailing list
>>>>>>>> Neuroimaging at python.org
>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Neuroimaging mailing list
>>>>>>> Neuroimaging at python.org
>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Neuroimaging mailing list
>>>>>> Neuroimaging at python.org
>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160531/f30358d6/attachment-0001.html>

From garyfallidis at gmail.com  Tue May 31 18:39:11 2016
From: garyfallidis at gmail.com (Eleftherios Garyfallidis)
Date: Tue, 31 May 2016 22:39:11 +0000
Subject: [Neuroimaging] Fwd: [dipy] Issues trying to install dipy
In-Reply-To: <CAAKhGh-y=eMep32OYsrEyr_XhxKD_9p4RVBgK5RAu6CxkRiVag@mail.gmail.com>
References: <CAAKhGh_7FAJ6xeST+nD9ShVTpu7mpS5C=HSok85aMmECqPhT_w@mail.gmail.com>
 <CAAKhGh9-emozuiqQv2KpY8hghks-1b-bid1PQufJ7YUwDffoDQ@mail.gmail.com>
 <CAH6Pt5rVree2gyERp3DSXVmC_dS4j4gQ66YJpouO62W4JpvX=g@mail.gmail.com>
 <CA+X4w0zbG_+5FQx8ALCvij54y+p_4WyfpwfD3LyigWoxAqH66A@mail.gmail.com>
 <CAAKhGh_7yUqHe7EVVmpNvNSbZuv1029CfjvK8Vb3Od57FRssEg@mail.gmail.com>
 <CA+X4w0z=1qjsbLcknNSjMZOkAjrSxhE=8B23E4+_YJPjcJTVrw@mail.gmail.com>
 <CAAKhGh800SY0ooqJe9_X1BVnOzbVqjGVtaYOHU-zHGLS1n+1jQ@mail.gmail.com>
 <CAN6JmWOueUqqsiBAHUm9tgyimNxdDUPgJsA7d2aU5d5WxFMzJg@mail.gmail.com>
 <CAAKhGh9x5Jhp78FQKaExxYSCAOBh3qxzVWfy+nfPskf0dA0Lnw@mail.gmail.com>
 <CAJR3sXd3ESkxq44qgaiKRPr_ROwQpT-sZTLrGcb7OGxjGOVLdg@mail.gmail.com>
 <CAAKhGh-y=eMep32OYsrEyr_XhxKD_9p4RVBgK5RAu6CxkRiVag@mail.gmail.com>
Message-ID: <CAN6JmWMaY8zfoX=8OLBQisBNwZydgGGwzf5hcg067jtL6btPCQ@mail.gmail.com>

Hi Jon,

When I develop DIPY using Anaconda in Windows I don't think I need to
install a compiler separately. Anaconda is coming with its own MinGW
compiler. Could it be that you are adding an extra compiler when it is not
necessary?

Also I 'll be happy to hangout and help you with your installation problem.
Send me an e-mail off the list please to arrange a meeting.

Cheers,
Eleftherios



On Tue, May 31, 2016 at 5:34 PM Jon Haitz Legarreta <
jhlegarreta at vicomtech.org> wrote:

> Hi Greg,
> using the virtual env I'm able to run the examples without major issues.
> Indeed, for the virtual env I downloaded the conda-forge dipy package.
>
> But my intent is to build from the latest source.
>
> Thanks,
> JON HAITZ
>
>
>
>
>
> On 31 May 2016 at 22:01, Gregory Lee <grlee77 at gmail.com> wrote:
>
>> Hi Jon,
>>
>> Do you need to build from the lastest source or are you just trying to
>> install a working version to run examples?  If the later, it may be worth
>> trying the pre-built conda packages Ariel recently created at conda-forge.
>> To install:
>>
>> conda install -c conda-forge dipy
>>
>> - Greg
>>
>> On Tue, May 31, 2016 at 2:58 PM, Jon Haitz Legarreta <
>> jhlegarreta at vicomtech.org> wrote:
>>
>>> Dear Eleftherios,
>>> thanks for the follow-up.
>>>
>>> No, I did not contact the Anaconda developers, since I was a little bit
>>> lost in the messages/source problem. But if your guess is that it is more
>>> likely an Anaconda problem, I will contact them and let you know.
>>>
>>> On the other hand, you are right; I installed Anaconda, then tried to
>>> set up dipy for development.
>>>
>>> Also I have a virtual environment where I downloaded the dipy
>>> dependencies. I do not know if the latter step can be avoided or else,
>>> whether one can be substituted by the other: i.e. and whether when one
>>> installs dipy for development from sources, dipy scripts take care of
>>> putting in place the necessary packages. I just thought that creating a
>>> virtual env with just the dipy dependencies would be cleaner for
>>> development (i.e. avoid clashes with other packages or repos, etc.)
>>>
>>> Since one of the links posted suggested it, I installed MinGW, but had
>>> no other compiler installer on my machine.
>>>
>>> Sincerely,
>>> JON HAITZ
>>>
>>>
>>>
>>>
>>> On 31 May 2016 at 00:16, Eleftherios Garyfallidis <
>>> garyfallidis at gmail.com> wrote:
>>>
>>>> Hi Jon,
>>>>
>>>> The error is not related to SSE or to OMP. Those are ommitted and then
>>>> the compilation continues properly. The problem appears later. Here is the
>>>> message
>>>>
>>>> C:\Anaconda3\libs/python35.lib: error adding symbols: File format not recognized
>>>> collect2.exe: error: ld returned 1 exit status
>>>>
>>>> Have you contacted the Anaconda developers? This looks like a problem
>>>> on their side.
>>>>
>>>> Let us know what they said to you. Otherwise I wonder if this is a
>>>> specific problem with Python 3 or if it affects also Python 2. You may want
>>>> to try that too. The problem does look more likely to be related to the
>>>> compiler used.
>>>>
>>>> Am I correct to say that the only thing that you did was to install
>>>> Anaconda and then pip install dipy? Did you have other compilers already
>>>> installed in your system?
>>>>
>>>> Best regards,
>>>> Eleftherios
>>>>
>>>> On Mon, May 30, 2016 at 5:51 PM Jon Haitz Legarreta <
>>>> jhlegarreta at vicomtech.org> wrote:
>>>>
>>>>> Dear Ariel,
>>>>> thanks for your suggestion.
>>>>>
>>>>> The patches in the link seem to help a little bit, but the process
>>>>> seems still to be unsuccessful: the MinGW gcc complains with the message:
>>>>> gcc: error: /arch:SSE2: No such file or directory
>>>>>
>>>>> Attached is the new log.
>>>>>
>>>>> Again, googling was not of much help. I got bits and parts of related
>>>>> errors, but have no clear picture of the issue.
>>>>>
>>>>> Any suggestion is appreciated.
>>>>>
>>>>> JON HAITZ
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 24 May 2016 at 01:08, Ariel Rokem <arokem at gmail.com> wrote:
>>>>>
>>>>>>
>>>>>>
>>>>>> On Mon, May 23, 2016 at 3:19 PM, Jon Haitz Legarreta <
>>>>>> jhlegarreta at vicomtech.org> wrote:
>>>>>>
>>>>>>> Hi there,
>>>>>>> thank you Matthew and Ariel.
>>>>>>>
>>>>>>> The link pointed by Ariel does not seem to be a solution; after
>>>>>>> having installed MinGW, as suggested in the link and although I'm aware it
>>>>>>> might be unnecessary, the Anaconda3 powershell still yields a similar
>>>>>>> error, now pointing to MSVC (which I do not have on my system):
>>>>>>>
>>>>>>>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 157, in
>>>>>>> __init__
>>>>>>>     self.dll_libraries = get_msvcr()
>>>>>>>   File "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 86, in
>>>>>>> get_msvcr
>>>>>>>     raise ValueError("Unknown MS Compiler version %s " % msc_ver)
>>>>>>> ValueError: Unknown MS Compiler version 1900
>>>>>>>
>>>>>>> Looks like maybe you ran into this corner case?
>>>>>>
>>>>>> http://stackoverflow.com/a/34427014/3532933
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>> I'll try to investigate further, and will let you know.
>>>>>>>
>>>>>>> Kind regards,
>>>>>>> JON HAITZ
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On 21 May 2016 at 16:55, Ariel Rokem <arokem at gmail.com> wrote:
>>>>>>>
>>>>>>>> Hi Jon and Matthew,
>>>>>>>>
>>>>>>>>
>>>>>>>> On Sat, May 21, 2016 at 7:30 AM, Matthew Brett <
>>>>>>>> matthew.brett at gmail.com> wrote:
>>>>>>>>
>>>>>>>>> Hi,
>>>>>>>>>
>>>>>>>>> On Sat, May 21, 2016 at 9:18 AM, Jon Haitz Legarreta
>>>>>>>>> <jhlegarreta at vicomtech.org> wrote:
>>>>>>>>> > Hi there,
>>>>>>>>> > has anybody experienced the issue below?
>>>>>>>>> >
>>>>>>>>> > Thanks,
>>>>>>>>> > JON HAITZ
>>>>>>>>> >
>>>>>>>>> >
>>>>>>>>> >
>>>>>>>>> >
>>>>>>>>> > ---------- Forwarded message ----------
>>>>>>>>> > From: Jon Haitz Legarreta <jhlegarreta at vicomtech.org>
>>>>>>>>> > Date: 18 May 2016 at 19:10
>>>>>>>>> > Subject: [dipy] Issues trying to install dipy
>>>>>>>>> > To: neuroimaging at python.org
>>>>>>>>> >
>>>>>>>>> >
>>>>>>>>> > Hi there,
>>>>>>>>> > I'm a newbie to dipy.
>>>>>>>>> >
>>>>>>>>> > I was trying to follow the instructions in [1] to have dipy
>>>>>>>>> installed from
>>>>>>>>> > the source code, so that I could execute the dipy examples.
>>>>>>>>> >
>>>>>>>>> > I'm using Windows 10 and Anaconda 3.
>>>>>>>>> >
>>>>>>>>> > When trying to execute
>>>>>>>>> > python setup.py develop
>>>>>>>>> >
>>>>>>>>> > the Anaconda prompt yields an error that says in the end:
>>>>>>>>> > File: "C:\Anaconda3\lib\distutils\cygwinccompiler.py", line 126,
>>>>>>>>> __init__
>>>>>>>>> > if self.ld_version >= "2.10.90":
>>>>>>>>> > TypeError: unorderable types: NoneType() >= str()
>>>>>>>>> >
>>>>>>>>> > I've been googling for a solution without success.
>>>>>>>>> >
>>>>>>>>> > I don't know whether this looks like Anaconda3 is trying to use
>>>>>>>>> cygwin
>>>>>>>>> > instead of mingw32, and whether that is the root cause.
>>>>>>>>> >
>>>>>>>>> > In either case, does anyone know how to solve the issue?
>>>>>>>>> >
>>>>>>>>> > Attached is the trace (it's short) of the error if this is of
>>>>>>>>> any help.
>>>>>>>>> >
>>>>>>>>> >
>>>>>>>>> > Thank you,
>>>>>>>>> > JON HAITZ
>>>>>>>>> >
>>>>>>>>> >
>>>>>>>>> > [1] http://nipy.org/dipy/installation.html#install-source-nix
>>>>>>>>>
>>>>>>>>> I'm sorry, I'm afraid I don't personally use Anaconda, so I have no
>>>>>>>>> experience of fixing compilation errors on Anaconda.   Ariel - have
>>>>>>>>> you come across this?
>>>>>>>>>
>>>>>>>>
>>>>>>>> And I don't personally use Windows...
>>>>>>>>
>>>>>>>> Might this be helpful:
>>>>>>>>
>>>>>>>>
>>>>>>>> http://stackoverflow.com/questions/24683305/python-cant-install-packages-typeerror-unorderable-types-nonetype-str
>>>>>>>>
>>>>>>>> It seems like it could be related, though it's all Greek to me.
>>>>>>>>
>>>>>>>> Cheers,
>>>>>>>>
>>>>>>>> Ariel
>>>>>>>>
>>>>>>>>
>>>>>>>>> You could also try on the anaconda support channels (issues,
>>>>>>>>> mailing
>>>>>>>>> list) - it may well be a general problem rather than one specific
>>>>>>>>> to
>>>>>>>>> dipy,
>>>>>>>>>
>>>>>>>>> Best,
>>>>>>>>>
>>>>>>>>> Matthew
>>>>>>>>> _______________________________________________
>>>>>>>>> Neuroimaging mailing list
>>>>>>>>> Neuroimaging at python.org
>>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> Neuroimaging mailing list
>>>>>>>> Neuroimaging at python.org
>>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Neuroimaging mailing list
>>>>>>> Neuroimaging at python.org
>>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Neuroimaging mailing list
>>>>>> Neuroimaging at python.org
>>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>>
>>>>>>
>>>>> _______________________________________________
>>>>> Neuroimaging mailing list
>>>>> Neuroimaging at python.org
>>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>>
>>>>
>>>> _______________________________________________
>>>> Neuroimaging mailing list
>>>> Neuroimaging at python.org
>>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>>
>>>>
>>>
>>> _______________________________________________
>>> Neuroimaging mailing list
>>> Neuroimaging at python.org
>>> https://mail.python.org/mailman/listinfo/neuroimaging
>>>
>>>
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160531/86b764b6/attachment.html>