From wagnerfl at student.ethz.ch Fri Nov 1 05:42:10 2013 From: wagnerfl at student.ethz.ch (Florian M. Wagner) Date: Fri, 1 Nov 2013 10:42:10 +0100 Subject: [IPython-dev] Numpy with OpenBLAS vs. IPython In-Reply-To: <5270C753.30805@student.ethz.ch> References: <526FDFF5.30304@student.ethz.ch> <527000F0.20704@googlemail.com> <5270C465.1000107@student.ethz.ch> <5270C753.30805@student.ethz.ch> Message-ID: <52737772.7040104@student.ethz.ch> Any ideas? Has someone here a functioning setup with Numpy-OpenBlas and the scientific (I)Python stack on Linux and is willing to share the necessary installation steps? Am 30.10.2013 09:46, schrieb Florian M. Wagner: > I reproduced the error with the latest IPython source: > > fwagner at fwagner-notebook ~/Downloads/ipython $ ipython > Python 2.7.4 (default, Sep 26 2013, 03:20:26) > Type "copyright", "credits" or "license" for more information. > > IPython 2.0.0-dev -- An enhanced Interactive Python. > ? -> Introduction and overview of IPython's features. > %quickref -> Quick reference. > help -> Python's own help system. > object? -> Details about 'object', use 'object??' for extra details. > > In [1]: A = np.random.randn(50,50) > > In [2]: A[0,0] > Out[2]: 0.62268531821589967 > > In [3]: A.T.dot(A) > Speicherzugriffsfehler > > Compared to: > > Python 2.7.4 (default, Sep 26 2013, 03:20:26) > [GCC 4.7.3] on linux2 > Type "help", "copyright", "credits" or "license" for more information. > >>> import numpy as np > >>> A = np.random.randn(50,50) > >>> A.T.dot(A) > array([[ 4.61866860e+01, 1.97714469e+00, 8.78842173e+00, ..., > -9.65143259e+00, 4.74312877e+00, -1.85163266e+01], > [ 1.97714469e+00, 4.50113072e+01, -1.13105045e+01, ..., > 2.31604581e+00, 4.93068512e+00, 1.29316267e+01], > [ 8.78842173e+00, -1.13105045e+01, 6.02859864e+01, ..., > -2.65040891e+00, 3.98374437e+00, -1.54860323e+01], > ..., > [ -9.65143259e+00, 2.31604581e+00, -2.65040891e+00, ..., > 3.81774984e+01, 1.63768677e+00, 1.16681961e+01], > [ 4.74312877e+00, 4.93068512e+00, 3.98374437e+00, ..., > 1.63768677e+00, 5.11159138e+01, 6.56039753e-02], > [ -1.85163266e+01, 1.29316267e+01, -1.54860323e+01, ..., > 1.16681961e+01, 6.56039753e-02, 6.61061915e+01]]) > > > > Am 30.10.2013 09:33, schrieb Florian M. Wagner: >> Thanks guys, >> >> @Julian: Reducing the thread number did not help >> @Thomas: dap is mpl-basemap dependency, fixed the import issue, but the >> error still occurs >> >> It is really weird that it is occuring only when printing the array, or >> slicing the array. And whats more weird, that it works in plain Python >> as I understood that IPython has no numpy dependency? >> >> Thanks and regards >> Florian >> >> Am 29.10.2013 19:39, schrieb Julian Taylor: >>> does this also happen if you do this before starting ipython? >>> >>> export OPENBLAS_NUM_THREADS=1 >>> (assuming bash shell) >>> >>> On 29.10.2013 17:19, Florian M. Wagner wrote: >>>> Dear all, >>>> >>>> I recently compiled numpy with OpenBlas for faster matrix-matrix >>>> multiplication. Everything works fine, but IPython seems to dislike it. >>>> I get a Segmentation Fault error when trying this (I reinstalled IPython >>>> 1.1.0 after numpy compilation): >>>> >>>> fwagner at fwagner-notebook ~ $ ipython -c "import numpy as np; print >>>> np.__version__; import IPython; print IPython.__version__; A = >>>> np.random.randn(600, 600); print A[20,53]; print A" >>>> /usr/local/bin/ipython:5: UserWarning: Module dap was already >>>> imported from None, but /usr/lib/python2.7/dist-packages is being >>>> added to sys.path >>>> from pkg_resources import load_entry_point >>>> 1.9.0.dev-54d3559 >>>> 1.1.0 >>>> -2.1004055886 >>>> Speicherzugriffsfehler >>>> >>>> In comparison, this works fine (just python instead of ipython): >>>> >>>> fwagner at fwagner-notebook ~ $ python -c "import numpy as np; print >>>> np.__version__; import IPython; print IPython.__version__; A = >>>> np.random.randn(600, 600); print A[20,53]; print A" >>>> 1.9.0.dev-54d3559 >>>> 1.1.0 >>>> -1.10282239744 >>>> [[ 0.13798084 -0.39195796 -0.15309986 ..., 1.47471232 -0.9158455 >>>> 0.97060671] >>>> [ 0.25720169 0.51895478 0.55175787 ..., -0.02010573 0.78047754 >>>> -1.07891961] >>>> [ 1.59478156 0.81245486 -0.03959768 ..., -0.04816659 -0.03356907 >>>> -0.11860379] >>>> ..., >>>> [-0.06299817 -0.0956477 -0.32987518 ..., -1.47714386 1.66736916 >>>> -0.47704276] >>>> [-0.93769589 1.50006542 0.97782803 ..., -0.5034154 1.31592011 >>>> -0.74482995] >>>> [-0.24661336 -0.98979706 -0.39366363 ..., 0.20507632 0.8701609 >>>> -1.10449957]] >>>> >>>> Any ideas what might cause this problem? >>>> >>>> Thanks, >>>> >>>> Florian >>>> >>>> >>>> >>>> _______________________________________________ >>>> IPython-dev mailing list >>>> IPython-dev at scipy.org >>>> http://mail.scipy.org/mailman/listinfo/ipython-dev >>>> >> _______________________________________________ >> IPython-dev mailing list >> IPython-dev at scipy.org >> http://mail.scipy.org/mailman/listinfo/ipython-dev > > > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev -------------- next part -------------- An HTML attachment was scrubbed... URL: From jenshnielsen at gmail.com Fri Nov 1 06:01:41 2013 From: jenshnielsen at gmail.com (Jens Nielsen) Date: Fri, 1 Nov 2013 10:01:41 +0000 Subject: [IPython-dev] Numpy with OpenBLAS vs. IPython In-Reply-To: <52737772.7040104@student.ethz.ch> References: <526FDFF5.30304@student.ethz.ch> <527000F0.20704@googlemail.com> <5270C465.1000107@student.ethz.ch> <5270C753.30805@student.ethz.ch> <52737772.7040104@student.ethz.ch> Message-ID: Hi I have just tried this using openblas 0.2.8 and numpy 1.8.0 on mac OSx with ipython 1.1.0 and I can't reproduce your issue. What is the output of numpy.show_config() from python and IPython? Is the same install of numpy being picked up and used. I.e. what is the output of numpy.__file__ in python and IPython? best regards Jens On Fri, Nov 1, 2013 at 9:42 AM, Florian M. Wagner wrote: > Any ideas? Has someone here a functioning setup with Numpy-OpenBlas and > the scientific (I)Python stack on Linux and is willing to share the > necessary installation steps? > > Am 30.10.2013 09:46, schrieb Florian M. Wagner: > > I reproduced the error with the latest IPython source: > > fwagner at fwagner-notebook ~/Downloads/ipython $ ipython > Python 2.7.4 (default, Sep 26 2013, 03:20:26) > Type "copyright", "credits" or "license" for more information. > > IPython 2.0.0-dev -- An enhanced Interactive Python. > ? -> Introduction and overview of IPython's features. > %quickref -> Quick reference. > help -> Python's own help system. > object? -> Details about 'object', use 'object??' for extra details. > > In [1]: A = np.random.randn(50,50) > > In [2]: A[0,0] > Out[2]: 0.62268531821589967 > > In [3]: A.T.dot(A) > Speicherzugriffsfehler > > Compared to: > > Python 2.7.4 (default, Sep 26 2013, 03:20:26) > [GCC 4.7.3] on linux2 > Type "help", "copyright", "credits" or "license" for more information. > >>> import numpy as np > >>> A = np.random.randn(50,50) > >>> A.T.dot(A) > array([[ 4.61866860e+01, 1.97714469e+00, 8.78842173e+00, ..., > -9.65143259e+00, 4.74312877e+00, -1.85163266e+01], > [ 1.97714469e+00, 4.50113072e+01, -1.13105045e+01, ..., > 2.31604581e+00, 4.93068512e+00, 1.29316267e+01], > [ 8.78842173e+00, -1.13105045e+01, 6.02859864e+01, ..., > -2.65040891e+00, 3.98374437e+00, -1.54860323e+01], > ..., > [ -9.65143259e+00, 2.31604581e+00, -2.65040891e+00, ..., > 3.81774984e+01, 1.63768677e+00, 1.16681961e+01], > [ 4.74312877e+00, 4.93068512e+00, 3.98374437e+00, ..., > 1.63768677e+00, 5.11159138e+01, 6.56039753e-02], > [ -1.85163266e+01, 1.29316267e+01, -1.54860323e+01, ..., > 1.16681961e+01, 6.56039753e-02, 6.61061915e+01]]) > > > > Am 30.10.2013 09:33, schrieb Florian M. Wagner: > > Thanks guys, > > @Julian: Reducing the thread number did not help > @Thomas: dap is mpl-basemap dependency, fixed the import issue, but the > error still occurs > > It is really weird that it is occuring only when printing the array, or > slicing the array. And whats more weird, that it works in plain Python > as I understood that IPython has no numpy dependency? > > Thanks and regards > Florian > > Am 29.10.2013 19:39, schrieb Julian Taylor: > > does this also happen if you do this before starting ipython? > > export OPENBLAS_NUM_THREADS=1 > (assuming bash shell) > > On 29.10.2013 17:19, Florian M. Wagner wrote: > > Dear all, > > I recently compiled numpy with OpenBlas for faster matrix-matrix > multiplication. Everything works fine, but IPython seems to dislike it. > I get a Segmentation Fault error when trying this (I reinstalled IPython > 1.1.0 after numpy compilation): > > fwagner at fwagner-notebook ~ $ ipython -c "import numpy as np; print > np.__version__; import IPython; print IPython.__version__; A = > np.random.randn(600, 600); print A[20,53]; print A" > /usr/local/bin/ipython:5: UserWarning: Module dap was already > imported from None, but /usr/lib/python2.7/dist-packages is being > added to sys.path > from pkg_resources import load_entry_point > 1.9.0.dev-54d3559 > 1.1.0 > -2.1004055886 > Speicherzugriffsfehler > > In comparison, this works fine (just python instead of ipython): > > fwagner at fwagner-notebook ~ $ python -c "import numpy as np; print > np.__version__; import IPython; print IPython.__version__; A = > np.random.randn(600, 600); print A[20,53]; print A" > 1.9.0.dev-54d3559 > 1.1.0 > -1.10282239744 > [[ 0.13798084 -0.39195796 -0.15309986 ..., 1.47471232 -0.9158455 > 0.97060671] > [ 0.25720169 0.51895478 0.55175787 ..., -0.02010573 0.78047754 > -1.07891961] > [ 1.59478156 0.81245486 -0.03959768 ..., -0.04816659 -0.03356907 > -0.11860379] > ..., > [-0.06299817 -0.0956477 -0.32987518 ..., -1.47714386 1.66736916 > -0.47704276] > [-0.93769589 1.50006542 0.97782803 ..., -0.5034154 1.31592011 > -0.74482995] > [-0.24661336 -0.98979706 -0.39366363 ..., 0.20507632 0.8701609 > -1.10449957]] > > Any ideas what might cause this problem? > > Thanks, > > Florian > > > > _______________________________________________ > IPython-dev mailing listIPython-dev at scipy.orghttp://mail.scipy.org/mailman/listinfo/ipython-dev > > _______________________________________________ > IPython-dev mailing listIPython-dev at scipy.orghttp://mail.scipy.org/mailman/listinfo/ipython-dev > > > > > _______________________________________________ > IPython-dev mailing listIPython-dev at scipy.orghttp://mail.scipy.org/mailman/listinfo/ipython-dev > > > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From wagnerfl at student.ethz.ch Fri Nov 1 08:40:31 2013 From: wagnerfl at student.ethz.ch (Florian M. Wagner) Date: Fri, 1 Nov 2013 13:40:31 +0100 Subject: [IPython-dev] Numpy with OpenBLAS vs. IPython In-Reply-To: References: <526FDFF5.30304@student.ethz.ch> <527000F0.20704@googlemail.com> <5270C465.1000107@student.ethz.ch> <5270C753.30805@student.ethz.ch> <52737772.7040104@student.ethz.ch> Message-ID: <5273A13F.6010100@student.ethz.ch> Thanks Jens, you were right. numpy.__file__ was not the same. It works now! Best regards Florian Am 01.11.2013 11:01, schrieb Jens Nielsen: > Hi > > I have just tried this using openblas 0.2.8 and numpy 1.8.0 on mac OSx > with ipython 1.1.0 and I can't reproduce your issue. > What is the output of numpy.show_config() from python and IPython? > > Is the same install of numpy being picked up and used. I.e. what is > the output of numpy.__file__ in python and IPython? > > best regards > Jens > > > > > On Fri, Nov 1, 2013 at 9:42 AM, Florian M. Wagner > > wrote: > > Any ideas? Has someone here a functioning setup with > Numpy-OpenBlas and the scientific (I)Python stack on Linux and is > willing to share the necessary installation steps? > > Am 30.10.2013 09:46, schrieb Florian M. Wagner: >> I reproduced the error with the latest IPython source: >> >> fwagner at fwagner-notebook ~/Downloads/ipython $ ipython >> Python 2.7.4 (default, Sep 26 2013, 03:20:26) >> Type "copyright", "credits" or "license" for more information. >> >> IPython 2.0.0-dev -- An enhanced Interactive Python. >> ? -> Introduction and overview of IPython's features. >> %quickref -> Quick reference. >> help -> Python's own help system. >> object? -> Details about 'object', use 'object??' for extra >> details. >> >> In [1]: A = np.random.randn(50,50) >> >> In [2]: A[0,0] >> Out[2]: 0.62268531821589967 >> >> In [3]: A.T.dot(A) >> Speicherzugriffsfehler >> >> Compared to: >> >> Python 2.7.4 (default, Sep 26 2013, 03:20:26) >> [GCC 4.7.3] on linux2 >> Type "help", "copyright", "credits" or "license" for more >> information. >> >>> import numpy as np >> >>> A = np.random.randn(50,50) >> >>> A.T.dot(A) >> array([[ 4.61866860e+01, 1.97714469e+00, 8.78842173e+00, ..., >> -9.65143259e+00, 4.74312877e+00, -1.85163266e+01], >> [ 1.97714469e+00, 4.50113072e+01, -1.13105045e+01, ..., >> 2.31604581e+00, 4.93068512e+00, 1.29316267e+01], >> [ 8.78842173e+00, -1.13105045e+01, 6.02859864e+01, ..., >> -2.65040891e+00, 3.98374437e+00, -1.54860323e+01], >> ..., >> [ -9.65143259e+00, 2.31604581e+00, -2.65040891e+00, ..., >> 3.81774984e+01, 1.63768677e+00, 1.16681961e+01], >> [ 4.74312877e+00, 4.93068512e+00, 3.98374437e+00, ..., >> 1.63768677e+00, 5.11159138e+01, 6.56039753e-02], >> [ -1.85163266e+01, 1.29316267e+01, -1.54860323e+01, ..., >> 1.16681961e+01, 6.56039753e-02, 6.61061915e+01]]) >> >> >> >> Am 30.10.2013 09:33, schrieb Florian M. Wagner: >>> Thanks guys, >>> >>> @Julian: Reducing the thread number did not help >>> @Thomas: dap is mpl-basemap dependency, fixed the import issue, but the >>> error still occurs >>> >>> It is really weird that it is occuring only when printing the array, or >>> slicing the array. And whats more weird, that it works in plain Python >>> as I understood that IPython has no numpy dependency? >>> >>> Thanks and regards >>> Florian >>> >>> Am29.10.2013 19:39, schrieb Julian Taylor: >>>> does this also happen if you do this before starting ipython? >>>> >>>> export OPENBLAS_NUM_THREADS=1 >>>> (assuming bash shell) >>>> >>>> On29.10.2013 17:19, Florian M. Wagner wrote: >>>>> Dear all, >>>>> >>>>> I recently compiled numpy with OpenBlas for faster matrix-matrix >>>>> multiplication. Everything works fine, but IPython seems to dislike it. >>>>> I get a Segmentation Fault error when trying this (I reinstalled IPython >>>>> 1.1.0 after numpy compilation): >>>>> >>>>> fwagner at fwagner-notebook ~ $ ipython -c "import numpy as np; print >>>>> np.__version__; import IPython; print IPython.__version__; A = >>>>> np.random.randn(600, 600); print A[20,53]; print A" >>>>> /usr/local/bin/ipython:5: UserWarning: Module dap was already >>>>> imported from None, but /usr/lib/python2.7/dist-packages is being >>>>> added to sys.path >>>>> from pkg_resources import load_entry_point >>>>> 1.9.0.dev-54d3559 >>>>> 1.1.0 >>>>> -2.1004055886 >>>>> Speicherzugriffsfehler >>>>> >>>>> In comparison, this works fine (just python instead of ipython): >>>>> >>>>> fwagner at fwagner-notebook ~ $ python -c "import numpy as np; print >>>>> np.__version__; import IPython; print IPython.__version__; A = >>>>> np.random.randn(600, 600); print A[20,53]; print A" >>>>> 1.9.0.dev-54d3559 >>>>> 1.1.0 >>>>> -1.10282239744 >>>>> [[ 0.13798084 -0.39195796 -0.15309986 ..., 1.47471232 -0.9158455 >>>>> 0.97060671] >>>>> [ 0.25720169 0.51895478 0.55175787 ..., -0.02010573 0.78047754 >>>>> -1.07891961] >>>>> [ 1.59478156 0.81245486 -0.03959768 ..., -0.04816659 -0.03356907 >>>>> -0.11860379] >>>>> ..., >>>>> [-0.06299817 -0.0956477 -0.32987518 ..., -1.47714386 1.66736916 >>>>> -0.47704276] >>>>> [-0.93769589 1.50006542 0.97782803 ..., -0.5034154 1.31592011 >>>>> -0.74482995] >>>>> [-0.24661336 -0.98979706 -0.39366363 ..., 0.20507632 0.8701609 >>>>> -1.10449957]] >>>>> >>>>> Any ideas what might cause this problem? >>>>> >>>>> Thanks, >>>>> >>>>> Florian >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> IPython-dev mailing list >>>>> IPython-dev at scipy.org >>>>> http://mail.scipy.org/mailman/listinfo/ipython-dev >>>>> >>> _______________________________________________ >>> IPython-dev mailing list >>> IPython-dev at scipy.org >>> http://mail.scipy.org/mailman/listinfo/ipython-dev >> >> >> >> _______________________________________________ >> IPython-dev mailing list >> IPython-dev at scipy.org >> http://mail.scipy.org/mailman/listinfo/ipython-dev > > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > > > > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev -------------- next part -------------- An HTML attachment was scrubbed... URL: From jakevdp at cs.washington.edu Fri Nov 1 18:30:36 2013 From: jakevdp at cs.washington.edu (Jacob Vanderplas) Date: Fri, 1 Nov 2013 15:30:36 -0700 Subject: [IPython-dev] IPython notebook citation? Message-ID: Hi, Is there a good citation for the IPython notebook environment? I'm aware of the standard citation info on the IPython website [1], but this reference pre-dates the notebook by a few years. I'm currently working on a paper which utilizes the notebook specifically, and would like to give proper credit! Thanks, Jake [1] http://ipython.org/citing.html From fperez.net at gmail.com Fri Nov 1 18:50:32 2013 From: fperez.net at gmail.com (Fernando Perez) Date: Fri, 1 Nov 2013 15:50:32 -0700 Subject: [IPython-dev] IPython notebook citation? In-Reply-To: References: Message-ID: Hey Jake, unfortunately not. We're way, way overdue on writing at least three papers, one on the overall system architecture, one on the notebook and one on the parallel details. But who has time to write papers when there's so much code to have fun with? ;) So for now please at least do cite that paper, even if it's a gross injustice to all the newer work and the many contributors who aren't getting credit. I feel terrible for that, and hopefully we'll remedy it before long. Cheers, f On Fri, Nov 1, 2013 at 3:30 PM, Jacob Vanderplas wrote: > Hi, > Is there a good citation for the IPython notebook environment? I'm > aware of the standard citation info on the IPython website [1], but > this reference pre-dates the notebook by a few years. I'm currently > working on a paper which utilizes the notebook specifically, and would > like to give proper credit! > Thanks, > Jake > > [1] http://ipython.org/citing.html > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > -- Fernando Perez (@fperez_org; http://fperez.org) fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) fernando.perez-at-berkeley: contact me here for any direct mail -------------- next part -------------- An HTML attachment was scrubbed... URL: From ra092767 at ime.unicamp.br Sun Nov 3 04:17:38 2013 From: ra092767 at ime.unicamp.br (Raniere Silva) Date: Sun, 3 Nov 2013 07:17:38 -0200 Subject: [IPython-dev] 2to3 for IPython Notebook Message-ID: <20131103091738.GD30587@buriti> Hi, I need to convert some Python2 `print` statements to Python3 `print` function in .ipynb files. `2to3` didn't work. What tool I can use? Thanks, Raniere From takowl at gmail.com Sun Nov 3 12:14:53 2013 From: takowl at gmail.com (Thomas Kluyver) Date: Sun, 3 Nov 2013 09:14:53 -0800 Subject: [IPython-dev] 2to3 for IPython Notebook In-Reply-To: <20131103091738.GD30587@buriti> References: <20131103091738.GD30587@buriti> Message-ID: On 3 November 2013 01:17, Raniere Silva wrote: > I need to convert some Python2 `print` statements to Python3 `print` > function > in .ipynb files. `2to3` didn't work. What tool I can use? > Copying my comment from the Software Carpentry Github thread here: https://github.com/swcarpentry/bc/issues/107 We don't have an automated conversion tool for notebooks ourselves - we don't curate enough notebooks ourselves to really need one. However, it should be feasible to write something that loads a notebook, runs 2to3 on each of the code cells, and saves it again. I'm happy to help build that. It's not entirely trivial, though: - Some changes should only be done once per notebook, e.g. adding from __future__ imports. - IPython special syntax, e.g. %magics might break the parser 2to3 is using. Thanks, Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From aron at ahmadia.net Sun Nov 3 13:37:51 2013 From: aron at ahmadia.net (Aron Ahmadia) Date: Sun, 3 Nov 2013 13:37:51 -0500 Subject: [IPython-dev] 2to3 for IPython Notebook In-Reply-To: References: <20131103091738.GD30587@buriti> Message-ID: Hi Thomas, Thanks for this. Just to avoid confusion, let's keep the discussion in one place, I propose the issue you linked, since I suspect Greg Wilson will want to way in. Cheers, Aron On Sun, Nov 3, 2013 at 12:14 PM, Thomas Kluyver wrote: > > On 3 November 2013 01:17, Raniere Silva wrote: > >> I need to convert some Python2 `print` statements to Python3 `print` >> function >> in .ipynb files. `2to3` didn't work. What tool I can use? >> > > > Copying my comment from the Software Carpentry Github thread here: > https://github.com/swcarpentry/bc/issues/107 > > We don't have an automated conversion tool for notebooks ourselves - we > don't curate enough notebooks ourselves to really need one. However, it > should be feasible to write something that loads a notebook, runs 2to3 on > each of the code cells, and saves it again. I'm happy to help build that. > It's not entirely trivial, though: > > - Some changes should only be done once per notebook, e.g. adding from > __future__ imports. > - IPython special syntax, e.g. %magics might break the parser 2to3 is > using. > > Thanks, > > Thomas > > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From takowl at gmail.com Tue Nov 5 15:08:24 2013 From: takowl at gmail.com (Thomas Kluyver) Date: Tue, 5 Nov 2013 12:08:24 -0800 Subject: [IPython-dev] Linux default config directory moved in master Message-ID: Another heads up about recent changes: on Linux, we have switched the default config directory back to ~/.ipython (it had been ~/.config/ipython for a few releases). We decided that having the same default across platforms was easier for users, and easier for us to explain, than following the XDG recommendations, which we weren't doing properly anyway. In most cases, this will be taken care of automatically when you upgrade - IPython will move your existing config to the new location. Older versions of IPython can also use that folder if it exists. The only case which will require manual intervention is if both directories already exist. IPython 2.0.dev will now print a warning in this case. To get rid of it, merge your config into ~/.ipython, and delete ~/.config/ipython. Thanks, Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From bussonniermatthias at gmail.com Tue Nov 5 16:00:23 2013 From: bussonniermatthias at gmail.com (Matthias BUSSONNIER) Date: Tue, 5 Nov 2013 22:00:23 +0100 Subject: [IPython-dev] [Warning] Tooltip keybinding changed. Message-ID: <907E8242-C523-462B-9C46-BCFF91C844B8@gmail.com> Hi list, Inspired by Thomas last mail, I warn you about another recent change in https://github.com/ipython/ipython/pull/4398 In NotebookApp, showing tooltip when pressing Tab should be deactivated, you now have to press Shift-Tab, advantage, it works if you select a token in the middle of a cell too :-) You might not get the effect immediately as browser caching of JS files is quite aggressive sometime, but don't be surprised when it append. Shift-tab was already existing on 1.0, so start re-wiring your muscle memory. -- Matthias Completion still bound to Tab, Indent-Dedent is Ctrl-], Ctrl-[ From zvoros at gmail.com Tue Nov 5 16:13:54 2013 From: zvoros at gmail.com (=?ISO-8859-1?Q?Zolt=E1n_V=F6r=F6s?=) Date: Tue, 05 Nov 2013 22:13:54 +0100 Subject: [IPython-dev] question about tree Message-ID: <52795F92.4070605@gmail.com> Hi all, I have noticed that when I try to upload two notebooks with the same name, but from different directories, only one of them persists in the notebook dashboard. It is not so surprising, given that the notebook tree displays only the names, but not the folders, so there is no way of distinguishing the two files. But is this the intended behaviour, or do I miss something? Thanks, Zolt?n From pi at berkeley.edu Tue Nov 5 17:44:27 2013 From: pi at berkeley.edu (Paul Ivanov) Date: Tue, 5 Nov 2013 14:44:27 -0800 Subject: [IPython-dev] question about tree In-Reply-To: <52795F92.4070605@gmail.com> References: <52795F92.4070605@gmail.com> Message-ID: <20131105224427.GI27034@HbI-OTOH.berkeley.edu> Zolt?n V?r?s, on 2013-11-05 22:13, wrote: > Hi all, > > I have noticed that when I try to upload two notebooks with the same > name, but from different directories, only one of them persists in the > notebook dashboard. It is not so surprising, given that the notebook > tree displays only the names, but not the folders, so there is no way of > distinguishing the two files. But is this the intended behaviour, or do > I miss something? When you upload a notebook with a filename identical to one that already exists in that directory, it simply overwrites the old file in the directory the notebook server is serving. We should warn users when this is going to be the case, I have opened an issue to do just that. https://github.com/ipython/ipython/issues/4494 best, -- _ / \ A* \^ - ,./ _.`\\ / \ / ,--.S \/ \ / `"~,_ \ \ __o ? _ \<,_ /:\ --(_)/-(_)----.../ | \ --------------.......J Paul Ivanov http://pirsquared.org From fperez.net at gmail.com Tue Nov 5 19:48:41 2013 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 5 Nov 2013 16:48:41 -0800 Subject: [IPython-dev] question about tree In-Reply-To: <20131105224427.GI27034@HbI-OTOH.berkeley.edu> References: <52795F92.4070605@gmail.com> <20131105224427.GI27034@HbI-OTOH.berkeley.edu> Message-ID: But note that what Zoltan is referring to is files coming from different directories, but identically named (easy to get with our UntitledX naming scheme). We don't disambiguate those in the UI. I think the answer to this is simply that the UI for multidir support is yet to be written, Zoltan :) Once we build a proper UI for this, these issues will be correctly handled. For now, you can just use the notebook the 'old fashioned way', opening a new one for another directory. The multidir capabilities that are there are really 'plumbing' for us to be able to build the user layer on top, but that's not done yet. Cheers, f On Tue, Nov 5, 2013 at 2:44 PM, Paul Ivanov wrote: > Zolt?n V?r?s, on 2013-11-05 22:13, wrote: > > Hi all, > > > > I have noticed that when I try to upload two notebooks with the same > > name, but from different directories, only one of them persists in the > > notebook dashboard. It is not so surprising, given that the notebook > > tree displays only the names, but not the folders, so there is no way of > > distinguishing the two files. But is this the intended behaviour, or do > > I miss something? > > When you upload a notebook with a filename identical to one that > already exists in that directory, it simply overwrites the old > file in the directory the notebook server is serving. We should > warn users when this is going to be the case, I have opened an > issue to do just that. > > https://github.com/ipython/ipython/issues/4494 > > best, > -- > _ > / \ > A* \^ - > ,./ _.`\\ / \ > / ,--.S \/ \ > / `"~,_ \ \ > __o ? > _ \<,_ /:\ > --(_)/-(_)----.../ | \ > --------------.......J > Paul Ivanov > http://pirsquared.org > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > -- Fernando Perez (@fperez_org; http://fperez.org) fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) fernando.perez-at-berkeley: contact me here for any direct mail -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Tue Nov 5 20:13:58 2013 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 5 Nov 2013 17:13:58 -0800 Subject: [IPython-dev] question about tree In-Reply-To: References: <52795F92.4070605@gmail.com> <20131105224427.GI27034@HbI-OTOH.berkeley.edu> Message-ID: Never mind, I misread Zoltan's email; I missed the 'upload' part. Sorry about the noise. On Tue, Nov 5, 2013 at 4:48 PM, Fernando Perez wrote: > But note that what Zoltan is referring to is files coming from different > directories, but identically named (easy to get with our UntitledX naming > scheme). We don't disambiguate those in the UI. > > I think the answer to this is simply that the UI for multidir support is > yet to be written, Zoltan :) Once we build a proper UI for this, these > issues will be correctly handled. > > For now, you can just use the notebook the 'old fashioned way', opening a > new one for another directory. The multidir capabilities that are there are > really 'plumbing' for us to be able to build the user layer on top, but > that's not done yet. > > Cheers, > > f > > > On Tue, Nov 5, 2013 at 2:44 PM, Paul Ivanov wrote: > >> Zolt?n V?r?s, on 2013-11-05 22:13, wrote: >> > Hi all, >> > >> > I have noticed that when I try to upload two notebooks with the same >> > name, but from different directories, only one of them persists in the >> > notebook dashboard. It is not so surprising, given that the notebook >> > tree displays only the names, but not the folders, so there is no way of >> > distinguishing the two files. But is this the intended behaviour, or do >> > I miss something? >> >> When you upload a notebook with a filename identical to one that >> already exists in that directory, it simply overwrites the old >> file in the directory the notebook server is serving. We should >> warn users when this is going to be the case, I have opened an >> issue to do just that. >> >> https://github.com/ipython/ipython/issues/4494 >> >> best, >> -- >> _ >> / \ >> A* \^ - >> ,./ _.`\\ / \ >> / ,--.S \/ \ >> / `"~,_ \ \ >> __o ? >> _ \<,_ /:\ >> --(_)/-(_)----.../ | \ >> --------------.......J >> Paul Ivanov >> http://pirsquared.org >> _______________________________________________ >> IPython-dev mailing list >> IPython-dev at scipy.org >> http://mail.scipy.org/mailman/listinfo/ipython-dev >> > > > > -- > Fernando Perez (@fperez_org; http://fperez.org) > fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) > fernando.perez-at-berkeley: contact me here for any direct mail > -- Fernando Perez (@fperez_org; http://fperez.org) fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) fernando.perez-at-berkeley: contact me here for any direct mail -------------- next part -------------- An HTML attachment was scrubbed... URL: From pi at berkeley.edu Tue Nov 5 20:22:06 2013 From: pi at berkeley.edu (Paul Ivanov) Date: Tue, 5 Nov 2013 17:22:06 -0800 Subject: [IPython-dev] question about tree In-Reply-To: References: <52795F92.4070605@gmail.com> <20131105224427.GI27034@HbI-OTOH.berkeley.edu> Message-ID: <20131106012206.GJ27034@HbI-OTOH.berkeley.edu> Fernando Perez, on 2013-11-05 16:48, wrote: > But note that what Zoltan is referring to is files coming from different > directories, but identically named (easy to get with our UntitledX naming > scheme). We don't disambiguate those in the UI. It's not that we don't disambiguate them in the UI, it's that we only allow for one of those files to exist. When you upload any number of files, even from the same directory (because the dashboard doesn't keep track of where those files are coming from), if you give them identical names, you just keep overwriting the same file that corresponds to wherever the dashboard view is pointing to. That's because we allow users to upload to a specific directory: whichever directory is currently active and open in the dashboard view. The rest of Fernando's message is spot on, though. -- _ / \ A* \^ - ,./ _.`\\ / \ / ,--.S \/ \ / `"~,_ \ \ __o ? _ \<,_ /:\ --(_)/-(_)----.../ | \ --------------.......J Paul Ivanov http://pirsquared.org From eyaldechter at gmail.com Tue Nov 5 20:57:56 2013 From: eyaldechter at gmail.com (Eyal Dechter) Date: Tue, 5 Nov 2013 20:57:56 -0500 Subject: [IPython-dev] (no subject) Message-ID: Hi all, I have a question about logging. Is there a way to log all the messages sent from the frontends to the kernel? Thanks, Eyal -------------- next part -------------- An HTML attachment was scrubbed... URL: From zvoros at gmail.com Wed Nov 6 03:09:17 2013 From: zvoros at gmail.com (=?ISO-8859-1?Q?Zolt=E1n_V=F6r=F6s?=) Date: Wed, 06 Nov 2013 09:09:17 +0100 Subject: [IPython-dev] (no subject) In-Reply-To: References: Message-ID: <5279F92D.3000809@gmail.com> Hi Eyal, I might have misunderstood your question, but the command history is already recorded an available. It is in history.sqlite in your profile directory. Try %hist? Cheers, Zolt?n On 06/11/13 02:57, Eyal Dechter wrote: > Hi all, I have a question about logging. Is there a way to log all the > messages sent from the frontends to the kernel? > > Thanks, > Eyal > > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev -------------- next part -------------- An HTML attachment was scrubbed... URL: From bussonniermatthias at gmail.com Wed Nov 6 03:13:46 2013 From: bussonniermatthias at gmail.com (Matthias BUSSONNIER) Date: Wed, 6 Nov 2013 09:13:46 +0100 Subject: [IPython-dev] (no subject) In-Reply-To: References: Message-ID: <03E72673-A234-4278-AC3B-4F9076A1A44E@gmail.com> Le 6 nov. 2013 ? 02:57, Eyal Dechter a ?crit : > Hi all, I have a question about logging. Is there a way to log all the messages sent from the frontends to the kernel? > run with --debug will print most of them to stout. I guess one can config a file-logger too. Last resort, you can have your own proxy-kernel that act as a MITM and dump/inspect messages as the same time. --debug set log level to logging.DEBUG (maximize logging output) --log-level= (Application.log_level) Default: 30 Choices: (0, 10, 20, 30, 40, 50, 'DEBUG', 'INFO', 'WARN', 'ERROR', 'CRITICAL') Set the log level by value or name. -- M From zvoros at gmail.com Wed Nov 6 03:17:27 2013 From: zvoros at gmail.com (=?ISO-8859-1?Q?Zolt=E1n_V=F6r=F6s?=) Date: Wed, 06 Nov 2013 09:17:27 +0100 Subject: [IPython-dev] question about tree In-Reply-To: References: <52795F92.4070605@gmail.com> <20131105224427.GI27034@HbI-OTOH.berkeley.edu> Message-ID: <5279FB17.3040902@gmail.com> On 06/11/13 01:48, Fernando Perez wrote: > But note that what Zoltan is referring to is files coming from > different directories, but identically named (easy to get with our > UntitledX naming scheme). We don't disambiguate those in the UI. I actually tried to upload a notebook from a backup directory, but the problem is the same as with UntitledX. > I think the answer to this is simply that the UI for multidir support > is yet to be written, Zoltan :) Once we build a proper UI for this, > these issues will be correctly handled. I take it as a hint, and I would be willing to work on the problem, but I also think that there are some considerations that I don't know about. My obvious solution would be to display the full path (or the path relative to the serving directory), but there must be a reason why this hasn't been done:) Cheers, Zolt?n From pablo at sifflez.org Fri Nov 8 04:51:00 2013 From: pablo at sifflez.org (Pablo Oliveira) Date: Fri, 08 Nov 2013 10:51:00 +0100 Subject: [IPython-dev] Citing IPython Notebook Message-ID: <527CB404.3010005@sifflez.org> Hi, I'm preparing the final version of a paper that links to an IPython Notebook to help reproduce its results. I would like to cite IPython. The IPython webpage recommends to cite F/P?rez, /BE/Granger, //IPython: A System for Interactive Scientific Computing. /But this paper predated the Notebook part of IPython. Is there a more recent publication, describing how IPython can be used to share reproducible results through the Notebook? Is it ok to cite the Sloan Grant proposal (F Perez, BE Granger /An Open Source Framework for Interactive, Collaborative and Reproducible Scientific Computing and Education/)? Thanks, Pablo -------------- next part -------------- An HTML attachment was scrubbed... URL: From pi at berkeley.edu Fri Nov 8 14:16:17 2013 From: pi at berkeley.edu (Paul Ivanov) Date: Fri, 8 Nov 2013 11:16:17 -0800 Subject: [IPython-dev] Citing IPython Notebook In-Reply-To: <527CB404.3010005@sifflez.org> References: <527CB404.3010005@sifflez.org> Message-ID: <20131108191617.GH8575@HbI-OTOH.berkeley.edu> Hi Pablo, Pablo Oliveira, on 2013-11-08 10:51, wrote: > Hi, > > I'm preparing the final version of a paper that links to an IPython > Notebook to help > reproduce its results. I would like to cite IPython. > > The IPython webpage recommends to cite F/P?rez, /BE/Granger, //IPython: > A System for Interactive Scientific Computing. > /But this paper predated the Notebook part of IPython. Is there a more > recent publication, describing how IPython can be used to share > reproducible results through the Notebook? Is it ok to cite the Sloan > Grant proposal (F Perez, BE Granger /An Open Source Framework for > Interactive, Collaborative and Reproducible Scientific Computing and > Education/)? "Until we sit down and write an actual paper, it'd be great to cite both" - Fernando best, -- _ / \ A* \^ - ,./ _.`\\ / \ / ,--.S \/ \ / `"~,_ \ \ __o ? _ \<,_ /:\ --(_)/-(_)----.../ | \ --------------.......J Paul Ivanov http://pirsquared.org From pablo at sifflez.org Fri Nov 8 14:28:35 2013 From: pablo at sifflez.org (Pablo Oliveira) Date: Fri, 08 Nov 2013 20:28:35 +0100 Subject: [IPython-dev] Citing IPython Notebook In-Reply-To: <20131108191617.GH8575@HbI-OTOH.berkeley.edu> References: <527CB404.3010005@sifflez.org> <20131108191617.GH8575@HbI-OTOH.berkeley.edu> Message-ID: <527D3B63.10408@sifflez.org> On Fri 08 Nov 2013 08:16:17 PM CET, Paul Ivanov wrote: > Hi Pablo, > > Pablo Oliveira, on 2013-11-08 10:51, wrote: >> Hi, >> >> I'm preparing the final version of a paper that links to an IPython >> Notebook to help >> reproduce its results. I would like to cite IPython. >> >> The IPython webpage recommends to cite F/P?rez, /BE/Granger, //IPython: >> A System for Interactive Scientific Computing. >> /But this paper predated the Notebook part of IPython. Is there a more >> recent publication, describing how IPython can be used to share >> reproducible results through the Notebook? Is it ok to cite the Sloan >> Grant proposal (F Perez, BE Granger /An Open Source Framework for >> Interactive, Collaborative and Reproducible Scientific Computing and >> Education/)? > > "Until we sit down and write an actual paper, it'd be great to > cite both" - Fernando Ok sure. Thanks, Pablo From jsw at fnal.gov Fri Nov 8 14:35:48 2013 From: jsw at fnal.gov (Jon Wilson) Date: Fri, 8 Nov 2013 13:35:48 -0600 Subject: [IPython-dev] Citing IPython Notebook In-Reply-To: <527D3B63.10408@sifflez.org> References: <527CB404.3010005@sifflez.org> <20131108191617.GH8575@HbI-OTOH.berkeley.edu> <527D3B63.10408@sifflez.org> Message-ID: <527D3D14.5000307@fnal.gov> On a related note, is there a list anywhere of papers citing IPython? I have one to add to the list if so: T. Aaltonen et al. (CDF Collaboration), Physical Review Letters 111, 182002 (2013). [arXiv:1306.2357] Regards, Jon From takowl at gmail.com Sat Nov 9 16:49:52 2013 From: takowl at gmail.com (Thomas Kluyver) Date: Sat, 9 Nov 2013 13:49:52 -0800 Subject: [IPython-dev] Discovering kernels/profiles Message-ID: I'm starting to think about how newly installed kernels can be discovered by IPython, and the links between kernels and profiles. At the moment, the way to use a non-default kernel (such as IJulia) is to specify its arguments in a config value. As a shortcut, you can create a profile with this setting, and then launch IPython using that profile (e.g. --profile=julia). However, we'd like to make it possible to start different kernels from one notebook server (and presumably from one Qt console). When a kernel is installed, there should be a simple way for it to tell IPython of its existence, and how to start it. A profile is a collection of settings, along with its own command history. There may well be more than one profile for a given kernel - e.g. we ship a few Python profiles that load different libraries on startup. In some cases, it may also make sense to have more than one kernel per profile: the default profile is currently used for both Python 3 and Python 2. In most cases, however, it probably makes sense for each profile to be associated with one kernel. So, I propose: - On installation, kernels should place a small json file in a specific directory. There will be a per-user directory (e.g. ~/.ipython/kernels) and a systemwide one (e.g. /usr/share/ipython/kernels) - IPython will offer a mechanism to open a notebook (or run kernels by other means) with a specific profile - The list of 'profiles' will include the names of kernels that don't yet have a matching profile. When one of these is selected, a matching profile will be created - so the first time you use the Haskell kernel, say, a profile_haskell profile folder is made. - The profiles can specify a kernel definition file, instead of including kernel details directly in config. This allows the kernel definition to be updated later, if e.g. command line arguments change. Does this sound sensible? Then there's the question of how to handle multiple kernels sharing a profile. We could allow a general mechanism for this, or special case the Python 2/3 kernels, or just drop the possibility and force the Python kernels to use separate profiles. Thanks, Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From benjaminrk at gmail.com Sat Nov 9 17:53:32 2013 From: benjaminrk at gmail.com (MinRK) Date: Sat, 9 Nov 2013 14:53:32 -0800 Subject: [IPython-dev] Discovering kernels/profiles In-Reply-To: References: Message-ID: I think that in general, we want to be able to specify the kernel on a per-notebook basis, so we definitely need to remove the current 1:1 relationship of profile to kernel. Some ipython-wide location where kernel providers can say "here I am" makes sense. On Sat, Nov 9, 2013 at 1:49 PM, Thomas Kluyver wrote: > I'm starting to think about how newly installed kernels can be discovered > by IPython, and the links between kernels and profiles. > > At the moment, the way to use a non-default kernel (such as IJulia) is to > specify its arguments in a config value. As a shortcut, you can create a > profile with this setting, and then launch IPython using that profile (e.g. > --profile=julia). However, we'd like to make it possible to start different > kernels from one notebook server (and presumably from one Qt console). When > a kernel is installed, there should be a simple way for it to tell IPython > of its existence, and how to start it. > > A profile is a collection of settings, along with its own command history. > There may well be more than one profile for a given kernel - e.g. we ship a > few Python profiles that load different libraries on startup. In some > cases, it may also make sense to have more than one kernel per profile: the > default profile is currently used for both Python 3 and Python 2. In most > cases, however, it probably makes sense for each profile to be associated > with one kernel. > > So, I propose: > - On installation, kernels should place a small json file in a specific > directory. There will be a per-user directory (e.g. ~/.ipython/kernels) and > a systemwide one (e.g. /usr/share/ipython/kernels) > - IPython will offer a mechanism to open a notebook (or run kernels by > other means) with a specific profile > - The list of 'profiles' will include the names of kernels that don't yet > have a matching profile. When one of these is selected, a matching profile > will be created - so the first time you use the Haskell kernel, say, a > profile_haskell profile folder is made. > - The profiles can specify a kernel definition file, instead of including > kernel details directly in config. This allows the kernel definition to be > updated later, if e.g. command line arguments change. > > Does this sound sensible? > > Then there's the question of how to handle multiple kernels sharing a > profile. We could allow a general mechanism for this, or special case the > Python 2/3 kernels, or just drop the possibility and force the Python > kernels to use separate profiles. > > Thanks, > Thomas > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From bussonniermatthias at gmail.com Sun Nov 10 05:19:29 2013 From: bussonniermatthias at gmail.com (Matthias BUSSONNIER) Date: Sun, 10 Nov 2013 11:19:29 +0100 Subject: [IPython-dev] Discovering kernels/profiles In-Reply-To: References: Message-ID: <30A1DCEF-E0FA-4768-B0FB-40ED4E8A4445@gmail.com> Le 9 nov. 2013 ? 22:49, Thomas Kluyver a ?crit : > I'm starting to think about how newly installed kernels can be discovered by IPython, and the links between kernels and profiles. > .... > > Does this sound sensible? But other things might need to change when you change kernels. Like syntax highlighting, execute_on_complete_block, logo in the notebook, default code_mirror_theme. So you actually do need to have a full config logic loading that change depending on the kernel. You cannot **just** create a profile on the fly. I think that the "include the names of kernels that don't yet have a matching profile" make no sense. Either the kernel is installed and ships with a default-profile (in whatever form) or it does not exist. Also I would be more in favor of a structure along : .ipython/profiles_default/ - config_*.json - IJulia/ -config.json -notebook-config.json - IHaskell/ -config.json -notebook-config.json - IRuby/ -config.json -notebook-config.json ... .ipython/profiles_foo/ - config_*.json - IJulia/ -config.json -notebook-config.json - IHaskell/ -config.json -notebook-config.json - IRuby/ -config.json -notebook-config.json ... -- M > > Then there's the question of how to handle multiple kernels sharing a profile. We could allow a general mechanism for this, or special case the Python 2/3 kernels, or just drop the possibility and force the Python kernels to use separate profiles. > > Thanks, > Thomas > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev From benjaminrk at gmail.com Sun Nov 10 14:40:32 2013 From: benjaminrk at gmail.com (MinRK) Date: Sun, 10 Nov 2013 11:40:32 -0800 Subject: [IPython-dev] Discovering kernels/profiles In-Reply-To: <30A1DCEF-E0FA-4768-B0FB-40ED4E8A4445@gmail.com> References: <30A1DCEF-E0FA-4768-B0FB-40ED4E8A4445@gmail.com> Message-ID: On Sun, Nov 10, 2013 at 2:19 AM, Matthias BUSSONNIER < bussonniermatthias at gmail.com> wrote: > > Le 9 nov. 2013 ? 22:49, Thomas Kluyver a ?crit : > > > I'm starting to think about how newly installed kernels can be > discovered by IPython, and the links between kernels and profiles. > > .... > > > > Does this sound sensible? > > But other things might need to change when you change kernels. Like > syntax highlighting, > execute_on_complete_block, logo in the notebook, default code_mirror_theme. > Yes, and we need to figure out how to address that without creating a profile for each, because we need to be able to use multiple kernel types with one server instance, which means one profile. > So you actually do need to have a full config logic loading that change > depending on the kernel. > You cannot **just** create a profile on the fly. > > I think that the "include the names of kernels that don't yet have a > matching profile" > make no sense. Either the kernel is installed and ships with a > default-profile (in whatever form) > or it does not exist. > Also I would be more in favor of a structure along : > > .ipython/profiles_default/ > - config_*.json > - IJulia/ > -config.json > -notebook-config.json > - IHaskell/ > -config.json > -notebook-config.json > - IRuby/ > -config.json > -notebook-config.json > ... > .ipython/profiles_foo/ > - config_*.json > - IJulia/ > -config.json > -notebook-config.json > - IHaskell/ > -config.json > -notebook-config.json > - IRuby/ > -config.json > -notebook-config.json > ... > > -- > M > > > > > > > > Then there's the question of how to handle multiple kernels sharing a > profile. We could allow a general mechanism for this, or special case the > Python 2/3 kernels, or just drop the possibility and force the Python > kernels to use separate profiles. > > > > Thanks, > > Thomas > > _______________________________________________ > > IPython-dev mailing list > > IPython-dev at scipy.org > > http://mail.scipy.org/mailman/listinfo/ipython-dev > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > -------------- next part -------------- An HTML attachment was scrubbed... URL: From takowl at gmail.com Sun Nov 10 15:32:46 2013 From: takowl at gmail.com (Thomas Kluyver) Date: Sun, 10 Nov 2013 12:32:46 -0800 Subject: [IPython-dev] Discovering kernels/profiles In-Reply-To: References: <30A1DCEF-E0FA-4768-B0FB-40ED4E8A4445@gmail.com> Message-ID: On 10 November 2013 11:40, MinRK wrote: > >> But other things might need to change when you change kernels. Like >> syntax highlighting, >> execute_on_complete_block, logo in the notebook, default >> code_mirror_theme. >> > > Yes, and we need to figure out how to address that without creating a > profile for each, > because we need to be able to use multiple kernel types with one server > instance, > which means one profile. > Or possibly the server needs to take some of the settings from the notebook's profile (i.e. what we currently think of as the kernel's profile) when it displays a notebook. Matthias > Also I would be more in favor of a structure along : > > .ipython/profiles_default/ > - config_*.json > - IJulia/ > -config.json > -notebook-config.json I don't much like this nested structure, or the idea of yet another layer of config files interacting. I suspect we need to think more clearly about the config system. The different things that you can configure fall into several categories: - Settings that belong with the user, e.g. extra toolbar buttons, or syntax highlighting colours. - Settings that belong with the notebook: which kernel to start, possibly style options (but how does this interact with the user's own style settings?) - Settings that belong with the kernel: how to start the kernel, syntax highlighting lexer, maybe the logo to display in the UI. - User settings about the kernel: run this code on startup, etc. We need to work out which settings are in each category, where they should be stored, and how we should handle conflicts between them (e.g. if I have overridden the default style and the notebook also comes with a stylesheet, do we apply one or the other, or both?). Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefan at sun.ac.za Sun Nov 10 20:31:49 2013 From: stefan at sun.ac.za (=?iso-8859-1?Q?St=E9fan?= van der Walt) Date: Mon, 11 Nov 2013 03:31:49 +0200 Subject: [IPython-dev] Discovering kernels/profiles In-Reply-To: References: Message-ID: <20131111013149.GE18796@shinobi> On Sat, 09 Nov 2013 13:49:52 -0800, Thomas Kluyver wrote: > - On installation, kernels should place a small json file in a specific > directory. There will be a per-user directory (e.g. ~/.ipython/kernels) and > a systemwide one (e.g. /usr/share/ipython/kernels) It might also be good to expose an API to dynamically add kernels after startup (or at least keep that possibility in mind, so that the file approach doesn't lead to a design that favors configuring everything up-front). St?fan From takowl at gmail.com Sun Nov 10 21:03:39 2013 From: takowl at gmail.com (Thomas Kluyver) Date: Sun, 10 Nov 2013 18:03:39 -0800 Subject: [IPython-dev] Discovering kernels/profiles In-Reply-To: <20131111013149.GE18796@shinobi> References: <20131111013149.GE18796@shinobi> Message-ID: On 10 November 2013 17:31, St?fan van der Walt wrote: > It might also be good to expose an API to dynamically add kernels after > startup (or at least keep that possibility in mind, so that the file > approach doesn't lead to a design that favors configuring everything > up-front). > We can certainly keep it in mind, but I imagine installing a kernel is something people will do relatively rarely. So it may be that you have to manually trigger a rescan of the kernel definition files in order to see new kernel types after launching the server. Depending on how expensive that is, it could be triggered by reloading the dashboard, or it could be a separate handler. To do it totally dynamically, you'd need some kind of interprocess communication running from installers, and that sounds like a bad idea. Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From p.sopasakis at gmail.com Sun Nov 10 22:47:46 2013 From: p.sopasakis at gmail.com (Pantelis Sopasakis) Date: Mon, 11 Nov 2013 04:47:46 +0100 Subject: [IPython-dev] ipython-notebook on remote server peculiarity Message-ID: <99AC0D16-E078-4646-A69C-D0475CAAD6B1@gmail.com> Dear all, I posted a question related to python-notebook on stackoverflow, please see: http://stackoverflow.com/questions/19896119/ipython-notebook-on-remote-server-peculiarity I want to run an instance of ipython-notebook on a remote server and to access it remotely on the client's browser. For the time being I am not interested in security issues. I connect to the remote server over SSH and I start ipython-notebook using the commands: screen -S ipy ipython notebook --ip='*' ---pylab=inline --port=7777 So far so good - everything works excellently server-side and client-side (not only on my computer, but also from other clients). I then detach the screen (still everything OK) and then I log out from SSH. Then, I can connect to the remote server at http://someserver.sth:7777, but I can neither see the notebooks that I previously created listed, nor can I execute any code. Instead of my notebooks, I can only see a notebook titled 'Untitled0'. If I try to start my 'default' cluster, it gives me the error message: 'error starting cluster'. Server-side, I get the following error message (I uploaded it on github for your convenience): https://gist.github.com/alphaville/7407447 The strange this is that when I reconnect everything works smoothly again, and in particular I am receiving the following log messages: [NotebookApp] Using existing kernel: 1d102222-d480-4543-a0d9-628599fc3e6b [NotebookApp] Connecting to: tcp://127.0.0.1:44393 [NotebookApp] Connecting to: tcp://127.0.0.1:50684 [NotebookApp] Connecting to: tcp://127.0.0.1:45910 [NotebookApp] Using existing profile dir: u'/home/chung/.ipython/profile_default' [NotebookApp] Changing to working dir: /home/chung (the last line may be of particular interest). Of course keeping an open SSH connection to the server is not a solution? Could someone provide some hint? Any help is greatly appreciated. Best regards, Pantelis Sopasakis -------------- next part -------------- An HTML attachment was scrubbed... URL: From jabooth at gmail.com Mon Nov 11 03:21:18 2013 From: jabooth at gmail.com (James Booth) Date: Mon, 11 Nov 2013 08:21:18 +0000 Subject: [IPython-dev] ipython-notebook on remote server peculiarity In-Reply-To: <99AC0D16-E078-4646-A69C-D0475CAAD6B1@gmail.com> References: <99AC0D16-E078-4646-A69C-D0475CAAD6B1@gmail.com> Message-ID: Looks like the problem is that you don't have permission to access your home dir after logout: OSError: [Errno 13] Permission denied: '/home/chung/.ipython' Is your home drive encrypted? If so that could definitely be the problem, see here: http://askubuntu.com/questions/191142/is-it-possible-to-use-the-screen-command-to-work-with-files-on-an-encrypted-ho Best wishes James On 11 November 2013 03:47, Pantelis Sopasakis wrote: > Dear all, > > I posted a question related to python-notebook on stackoverflow, please > see: > > > http://stackoverflow.com/questions/19896119/ipython-notebook-on-remote-server-peculiarity > > I want to run an instance of ipython-notebook on a remote server and to > access it remotely on the client's browser. For the time being I am not > interested in security issues. > > I connect to the remote server over SSH and I start ipython-notebook using > the commands: > > screen -S ipy > ipython notebook --ip='*' ---pylab=inline --port=7777 > > So far so good - everything works excellently server-side and client-side > (not only on my computer, but also from other clients). > > I then detach the screen (still everything OK) and then I log out from > SSH. Then, I can connect to the remote server at > http://someserver.sth:7777, but I can neither see the notebooks that I > previously created listed, nor can I execute any code. Instead of my > notebooks, I can only see a notebook titled 'Untitled0'. If I try to start > my 'default' cluster, it gives me the error message: 'error starting > cluster'. Server-side, I get the following error message (I uploaded it on > github for your convenience): > > https://gist.github.com/alphaville/7407447 > > The strange this is that when I reconnect everything works smoothly again, > and in particular I am receiving the following log messages: > > [NotebookApp] Using existing kernel: 1d102222-d480-4543-a0d9-628599fc3e6b > [NotebookApp] Connecting to: tcp://127.0.0.1:44393 > [NotebookApp] Connecting to: tcp://127.0.0.1:50684 > [NotebookApp] Connecting to: tcp://127.0.0.1:45910 > [NotebookApp] Using existing profile dir: > u'/home/chung/.ipython/profile_default' > [NotebookApp] Changing to working dir: /home/chung > > (the last line may be of particular interest). > > Of course keeping an open SSH connection to the server is not a solution? > > Could someone provide some hint? Any help is greatly appreciated. > > Best regards, > Pantelis Sopasakis > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From p.sopasakis at gmail.com Mon Nov 11 11:56:59 2013 From: p.sopasakis at gmail.com (Pantelis Sopasakis) Date: Mon, 11 Nov 2013 17:56:59 +0100 Subject: [IPython-dev] ipython-notebook on remote server peculiarity In-Reply-To: References: <99AC0D16-E078-4646-A69C-D0475CAAD6B1@gmail.com> Message-ID: Dear James, Thank you very much for your answer. Indeed, my home directory on the sever is encrypted. I opted for a very easy solution: I just created a new user without encrypting its home directory. Everything is running smoothly now. Best, Pantelis On 11/nov/2013, at 09:21, James Booth wrote: > Looks like the problem is that you don't have permission to access your home dir after logout: > > OSError: [Errno 13] Permission denied: '/home/chung/.ipython' > > Is your home drive encrypted? If so that could definitely be the problem, see here: > > http://askubuntu.com/questions/191142/is-it-possible-to-use-the-screen-command-to-work-with-files-on-an-encrypted-ho > > Best wishes > James > > > > On 11 November 2013 03:47, Pantelis Sopasakis wrote: > Dear all, > > I posted a question related to python-notebook on stackoverflow, please see: > > http://stackoverflow.com/questions/19896119/ipython-notebook-on-remote-server-peculiarity > > I want to run an instance of ipython-notebook on a remote server and to access it remotely on the client's browser. For the time being I am not interested in security issues. > > I connect to the remote server over SSH and I start ipython-notebook using the commands: > > screen -S ipy > ipython notebook --ip='*' ---pylab=inline --port=7777 > > So far so good - everything works excellently server-side and client-side (not only on my computer, but also from other clients). > > I then detach the screen (still everything OK) and then I log out from SSH. Then, I can connect to the remote server at http://someserver.sth:7777, but I can neither see the notebooks that I previously created listed, nor can I execute any code. Instead of my notebooks, I can only see a notebook titled 'Untitled0'. If I try to start my 'default' cluster, it gives me the error message: 'error starting cluster'. Server-side, I get the following error message (I uploaded it on github for your convenience): > > https://gist.github.com/alphaville/7407447 > > The strange this is that when I reconnect everything works smoothly again, and in particular I am receiving the following log messages: > > [NotebookApp] Using existing kernel: 1d102222-d480-4543-a0d9-628599fc3e6b > [NotebookApp] Connecting to: tcp://127.0.0.1:44393 > [NotebookApp] Connecting to: tcp://127.0.0.1:50684 > [NotebookApp] Connecting to: tcp://127.0.0.1:45910 > [NotebookApp] Using existing profile dir: u'/home/chung/.ipython/profile_default' > [NotebookApp] Changing to working dir: /home/chung > > (the last line may be of particular interest). > > Of course keeping an open SSH connection to the server is not a solution? > > Could someone provide some hint? Any help is greatly appreciated. > > Best regards, > Pantelis Sopasakis > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev -------------- next part -------------- An HTML attachment was scrubbed... URL: From doc at doconnel.f9.co.uk Mon Nov 11 18:41:27 2013 From: doc at doconnel.f9.co.uk (Derek O'Connell) Date: Mon, 11 Nov 2013 23:41:27 +0000 Subject: [IPython-dev] Multi-Dashboard Message-ID: <52816B27.90805@doconnel.f9.co.uk> I'm running multiple local NB servers and want to see all the dashboards in one page. The obvious thing to do is to create a notebook using iframes to contain each dashboard. Seems to work but should I expect any problems? Here's the code I'm using: from IPython.display import HTML iframe = '
' ports = [8888,8889] frames = ''.join(iframe.format(p) for p in ports) HTML(frames) From bussonniermatthias at gmail.com Tue Nov 12 02:02:17 2013 From: bussonniermatthias at gmail.com (Matthias BUSSONNIER) Date: Tue, 12 Nov 2013 08:02:17 +0100 Subject: [IPython-dev] Multi-Dashboard In-Reply-To: <52816B27.90805@doconnel.f9.co.uk> References: <52816B27.90805@doconnel.f9.co.uk> Message-ID: Le 12 nov. 2013 ? 00:41, Derek O'Connell a ?crit : > I'm running multiple local NB servers and want to see all the dashboards > in one page. The obvious thing to do is to create a notebook using > iframes to contain each dashboard. Seems to work but should I expect any > problems? No, that should work. We'll probably get a way to navigate border in UI by december and 2.0. > Here's the code I'm using: > > from IPython.display import HTML Why not import IFrame ? :-) -- M > > iframe = '
' > ports = [8888,8889] > > frames = ''.join(iframe.format(p) for p in ports) > HTML(frames) > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev From doc at doconnel.f9.co.uk Tue Nov 12 02:26:09 2013 From: doc at doconnel.f9.co.uk (Derek O'Connell) Date: Tue, 12 Nov 2013 07:26:09 +0000 Subject: [IPython-dev] Multi-Dashboard In-Reply-To: References: <52816B27.90805@doconnel.f9.co.uk> Message-ID: <5281D811.6010904@doconnel.f9.co.uk> On 12/11/13 07:02, Matthias BUSSONNIER wrote: > > Le 12 nov. 2013 ? 00:41, Derek O'Connell a ?crit : > > > I'm running multiple local NB servers and want to see all the > > dashboards in one page. The obvious thing to do is to create a > > notebook using iframes to contain each dashboard. Seems to work but > > should I expect any problems? > > No, that should work. We'll probably get a way to navigate border in > UI by december and 2.0. Good to know! > > > > Here's the code I'm using: > > > > from IPython.display import HTML > > Why not import IFrame ? :-) Given the amount of white space in the dashboard I'm now putting two on a row using: iframe = '' Wouldn't IFrame restrict me to one per row/output? -D From bussonniermatthias at gmail.com Tue Nov 12 04:55:22 2013 From: bussonniermatthias at gmail.com (Matthias BUSSONNIER) Date: Tue, 12 Nov 2013 10:55:22 +0100 Subject: [IPython-dev] Multi-Dashboard In-Reply-To: <5281D811.6010904@doconnel.f9.co.uk> References: <52816B27.90805@doconnel.f9.co.uk> <5281D811.6010904@doconnel.f9.co.uk> Message-ID: Le 12 nov. 2013 ? 08:26, Derek O'Connell a ?crit : > On 12/11/13 07:02, Matthias BUSSONNIER wrote: >> >> Le 12 nov. 2013 ? 00:41, Derek O'Connell a ?crit : >> >>> I'm running multiple local NB servers and want to see all the >>> dashboards in one page. The obvious thing to do is to create a >>> notebook using iframes to contain each dashboard. Seems to work but >>> should I expect any problems? >> >> No, that should work. We'll probably get a way to navigate border in >> UI by december and 2.0. > > Good to know! You can already sorta navigate by editing the URL. > > iframe = '' > > Wouldn't IFrame restrict me to one per row/output? Yes, that's true. -- M From p.f.moore at gmail.com Tue Nov 12 11:24:34 2013 From: p.f.moore at gmail.com (Paul Moore) Date: Tue, 12 Nov 2013 16:24:34 +0000 Subject: [IPython-dev] Dependency on pyreadline Message-ID: On Windows, the IPython documentation says that pyreadline is a dependency. (From the website, "As indicated above, on Windows, PyReadline is a *mandatory* dependency".) However, I have on occasion seen problems with pyreadline, and it's not easy to disable it when that happens. So just to be clear - does anything in IPython *need* pyreadline? Other than the obvious improved command line experience (although to be honest, I find the basic Windows console behaviour fairly acceptable, a readline-type solution is by no means as crucial as it is on Unix). I've used IPython in the past without pyreadline, and there's nothing obvious broken. Thanks, Paul From takowl at gmail.com Tue Nov 12 20:19:21 2013 From: takowl at gmail.com (Thomas Kluyver) Date: Tue, 12 Nov 2013 17:19:21 -0800 Subject: [IPython-dev] Dependency on pyreadline In-Reply-To: References: Message-ID: On 12 November 2013 08:24, Paul Moore wrote: > On Windows, the IPython documentation says that pyreadline is a > dependency. (From the website, "As indicated above, on Windows, > PyReadline is a *mandatory* dependency".) However, I have on occasion > seen problems with pyreadline, and it's not easy to disable it when > that happens. > > So just to be clear - does anything in IPython *need* pyreadline? > No, that language can probably be softened. It was fairly mandatory when IPython was just a terminal interface, because without the features it provided you lost a big part of the reason to use IPython at all. But it's less important with the new interfaces, and I think the code allows for readline not being importable. Do you want to do a quick PR against the docs? Thanks, Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Wed Nov 13 01:23:31 2013 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 12 Nov 2013 22:23:31 -0800 Subject: [IPython-dev] New Data Science Initiative, aka where has Fernando been hiding for the last year? Message-ID: Hi all, this is an email that I've been waiting for almost a year to be able to write, and finally today I can do it... Almost a year ago, we announced the Sloan grant to support IPython development, and it would be reasonable to imagine that this would mean I would become significantly more active on the project than even before, since now IPython would be officially a large part of my job. However, astute observers will have noticed quite the opposite: my contributions in code, github activity and mailing list traffic have actually gone down since that day, not up (in my meager defense, I still do a fair bit behind the scenes and face-to-face in Berkeley :) Obviously middle age decay and manager syndrome can probably account for much of that, but there was another reason. On the *very same day* that we got the Sloan grant, I was pulled into a competition for another grant that Josh Greenberg, the same Sloan program director who funds us, was running in collaboration with the Moore foundation. This was a large effort to select three US universities for an ambitious project involving data science, where open source computational tools would play a central role. This became a very significant and time consuming project, but today we've been able to publicly announce the outcome, during an event at the White House OSTP. Rather than repeating in this email everything, I'll just point to a blog post I wrote with the rest of the story: http://blog.fperez.org/2013/11/an-ambitious-experiment-in-data-science.html What does this mean for IPython? Hopefully only good things: open source, and IPython specifically, were an important ingredient of the Berkeley proposal, and I expect to build at the new Berkeley Institute for Data Science a "place for people like us". The scientific Python culture at Berkeley is rapidly growing, in no small part thanks to the work of the great team at the DLab (dlab.berkeley.edu) who picked up my early py4science effort and now run a very active community. There's multiple projects on campus that involve scientific Python, all of them open source, and I am sure we'll be able to leverage this initiative in multiple productive ways for IPython and the larger ecosystem. As for myself, I do hope that next year I'll have more bandwidth for technical work on IPython. I have been able to remain plugged in to all our design work, but I *really* like to code, and I hate to be so far away from it. I hope my responsibilities at BIDS will still leave some room for it. Finally, I really want to thank Brian, Min, Thomas, Paul, Matthias and the rest of the team. You guys done far more than your fair share of the work, effectively picking up all of my slack in the most generous way imaginable. It's hard to think of a better team to work with. Cheers, f -- Fernando Perez (@fperez_org; http://fperez.org) fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) fernando.perez-at-berkeley: contact me here for any direct mail -------------- next part -------------- An HTML attachment was scrubbed... URL: From p.f.moore at gmail.com Wed Nov 13 03:14:27 2013 From: p.f.moore at gmail.com (Paul Moore) Date: Wed, 13 Nov 2013 08:14:27 +0000 Subject: [IPython-dev] Dependency on pyreadline In-Reply-To: References: Message-ID: On 13 November 2013 01:19, Thomas Kluyver wrote: > No, that language can probably be softened. It was fairly mandatory when > IPython was just a terminal interface, because without the features it > provided you lost a big part of the reason to use IPython at all. But it's > less important with the new interfaces, and I think the code allows for > readline not being importable. Cool. I'm pretty clear on the user-visible features that pyreadline gives normally, if there's nothing hidden internally within IPython, I can make a proper decision on the costs/benefits for myself. (By the way, personally I find even terminal IPython immensely useful even without readline) > Do you want to do a quick PR against the docs? I'd be happy to, may take a couple of days to find some free time, but I'll sort something out. Paul. From bussonniermatthias at gmail.com Wed Nov 13 08:18:28 2013 From: bussonniermatthias at gmail.com (Matthias BUSSONNIER) Date: Wed, 13 Nov 2013 14:18:28 +0100 Subject: [IPython-dev] New Data Science Initiative, aka where has Fernando been hiding for the last year? In-Reply-To: References: Message-ID: Hi Fernando, This is great news, I saw all the time end energy you've put into that, and I'm really happy it paid off. Also don't worry, even if you fell like you are not contributing as much as some point in the past, feel reassured, you are still omnipresent for us. Looking forward to see how all this will evolve in the futures month, and thanks for taking time to keep us informed. Thanks -- Matthias Le 13 nov. 2013 ? 07:23, Fernando Perez a ?crit : > Hi all, > > this is an email that I've been waiting for almost a year to be able to write, and finally today I can do it? > > Almost a year ago, we announced the Sloan grant to support IPython development, and it would be reasonable to imagine that this would mean I would become significantly more active on the project than even before, since now IPython would be officially a large part of my job. However, astute observers will have noticed quite the opposite: my contributions in code, github activity and mailing list traffic have actually gone down since that day, not up (in my meager defense, I still do a fair bit behind the scenes and face-to-face in Berkeley :) > > Obviously middle age decay and manager syndrome can probably account for much of that, but there was another reason. On the *very same day* that we got the Sloan grant, I was pulled into a competition for another grant that Josh Greenberg, the same Sloan program director who funds us, was running in collaboration with the Moore foundation. This was a large effort to select three US universities for an ambitious project involving data science, where open source computational tools would play a central role. > > This became a very significant and time consuming project, but today we've been able to publicly announce the outcome, during an event at the White House OSTP. Rather than repeating in this email everything, I'll just point to a blog post I wrote with the rest of the story: > > http://blog.fperez.org/2013/11/an-ambitious-experiment-in-data-science.html > > What does this mean for IPython? Hopefully only good things: open source, and IPython specifically, were an important ingredient of the Berkeley proposal, and I expect to build at the new Berkeley Institute for Data Science a "place for people like us". The scientific Python culture at Berkeley is rapidly growing, in no small part thanks to the work of the great team at the DLab (dlab.berkeley.edu) who picked up my early py4science effort and now run a very active community. There's multiple projects on campus that involve scientific Python, all of them open source, and I am sure we'll be able to leverage this initiative in multiple productive ways for IPython and the larger ecosystem. > > As for myself, I do hope that next year I'll have more bandwidth for technical work on IPython. I have been able to remain plugged in to all our design work, but I *really* like to code, and I hate to be so far away from it. I hope my responsibilities at BIDS will still leave some room for it. > > Finally, I really want to thank Brian, Min, Thomas, Paul, Matthias and the rest of the team. You guys done far more than your fair share of the work, effectively picking up all of my slack in the most generous way imaginable. It's hard to think of a better team to work with. > > Cheers, > > f > > -- > Fernando Perez (@fperez_org; http://fperez.org) > fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) > fernando.perez-at-berkeley: contact me here for any direct mail > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev -------------- next part -------------- An HTML attachment was scrubbed... URL: From ellisonbg at gmail.com Wed Nov 13 11:29:58 2013 From: ellisonbg at gmail.com (Brian Granger) Date: Wed, 13 Nov 2013 11:29:58 -0500 Subject: [IPython-dev] New Data Science Initiative, aka where has Fernando been hiding for the last year? In-Reply-To: References: Message-ID: Fernando, Glad you can finally talk about this publicly! This is super exciting for you, UC Berkeley, our community and IPython. Congratulations on a good - and hard - work done. We have missed you on IPython :), but this was a worthy investment of your time and energy. Looking forward to where this leads... Cheers, Brian On Wed, Nov 13, 2013 at 1:23 AM, Fernando Perez wrote: > Hi all, > > this is an email that I've been waiting for almost a year to be able to > write, and finally today I can do it... > > Almost a year ago, we announced the Sloan grant to support IPython > development, and it would be reasonable to imagine that this would mean I > would become significantly more active on the project than even before, > since now IPython would be officially a large part of my job. However, > astute observers will have noticed quite the opposite: my contributions in > code, github activity and mailing list traffic have actually gone down since > that day, not up (in my meager defense, I still do a fair bit behind the > scenes and face-to-face in Berkeley :) > > Obviously middle age decay and manager syndrome can probably account for > much of that, but there was another reason. On the *very same day* that we > got the Sloan grant, I was pulled into a competition for another grant that > Josh Greenberg, the same Sloan program director who funds us, was running in > collaboration with the Moore foundation. This was a large effort to select > three US universities for an ambitious project involving data science, where > open source computational tools would play a central role. > > This became a very significant and time consuming project, but today we've > been able to publicly announce the outcome, during an event at the White > House OSTP. Rather than repeating in this email everything, I'll just point > to a blog post I wrote with the rest of the story: > > http://blog.fperez.org/2013/11/an-ambitious-experiment-in-data-science.html > > What does this mean for IPython? Hopefully only good things: open source, > and IPython specifically, were an important ingredient of the Berkeley > proposal, and I expect to build at the new Berkeley Institute for Data > Science a "place for people like us". The scientific Python culture at > Berkeley is rapidly growing, in no small part thanks to the work of the > great team at the DLab (dlab.berkeley.edu) who picked up my early py4science > effort and now run a very active community. There's multiple projects on > campus that involve scientific Python, all of them open source, and I am > sure we'll be able to leverage this initiative in multiple productive ways > for IPython and the larger ecosystem. > > As for myself, I do hope that next year I'll have more bandwidth for > technical work on IPython. I have been able to remain plugged in to all our > design work, but I *really* like to code, and I hate to be so far away from > it. I hope my responsibilities at BIDS will still leave some room for it. > > Finally, I really want to thank Brian, Min, Thomas, Paul, Matthias and the > rest of the team. You guys done far more than your fair share of the work, > effectively picking up all of my slack in the most generous way imaginable. > It's hard to think of a better team to work with. > > Cheers, > > f > > -- > Fernando Perez (@fperez_org; http://fperez.org) > fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) > fernando.perez-at-berkeley: contact me here for any direct mail > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > -- Brian E. Granger Cal Poly State University, San Luis Obispo bgranger at calpoly.edu and ellisonbg at gmail.com From mail at telenczuk.pl Wed Nov 13 14:53:36 2013 From: mail at telenczuk.pl (Bartosz) Date: Wed, 13 Nov 2013 20:53:36 +0100 Subject: [IPython-dev] New Data Science Initiative, aka where has Fernando been hiding for the last year? In-Reply-To: References: Message-ID: <5283D8C0.3030107@telenczuk.pl> Hi Fernando, That's great news! Congratulations! We have been organising a school on Python in (data) science (last time in Zurich: http://python.g-node.org). The school has been a great success, but obtaining funding for a general school without any specific field (like bioinformatic, neuroscience etc.) is rather difficult in Europe. Could you explain what are the main goals of the new project and what is role Python/Ipython will play? Cheers, Bartosz On 11/13/2013 07:23 AM, Fernando Perez wrote: > Hi all, > > this is an email that I've been waiting for almost a year to be able to > write, and finally today I can do it... > > Almost a year ago, we announced the Sloan grant to support IPython > development, and it would be reasonable to imagine that this would mean > I would become significantly more active on the project than even > before, since now IPython would be officially a large part of my job. > However, astute observers will have noticed quite the opposite: my > contributions in code, github activity and mailing list traffic have > actually gone down since that day, not up (in my meager defense, I still > do a fair bit behind the scenes and face-to-face in Berkeley :) > > Obviously middle age decay and manager syndrome can probably account for > much of that, but there was another reason. On the *very same day* that > we got the Sloan grant, I was pulled into a competition for another > grant that Josh Greenberg, the same Sloan program director who funds us, > was running in collaboration with the Moore foundation. This was a large > effort to select three US universities for an ambitious project > involving data science, where open source computational tools would play > a central role. > > This became a very significant and time consuming project, but today > we've been able to publicly announce the outcome, during an event at the > White House OSTP. Rather than repeating in this email everything, I'll > just point to a blog post I wrote with the rest of the story: > > http://blog.fperez.org/2013/11/an-ambitious-experiment-in-data-science.html > > What does this mean for IPython? Hopefully only good things: open > source, and IPython specifically, were an important ingredient of the > Berkeley proposal, and I expect to build at the new Berkeley Institute > for Data Science a "place for people like us". The scientific Python > culture at Berkeley is rapidly growing, in no small part thanks to the > work of the great team at the DLab (dlab.berkeley.edu > ) who picked up my early py4science effort and > now run a very active community. There's multiple projects on campus > that involve scientific Python, all of them open source, and I am sure > we'll be able to leverage this initiative in multiple productive ways > for IPython and the larger ecosystem. > > As for myself, I do hope that next year I'll have more bandwidth for > technical work on IPython. I have been able to remain plugged in to all > our design work, but I *really* like to code, and I hate to be so far > away from it. I hope my responsibilities at BIDS will still leave some > room for it. > > Finally, I really want to thank Brian, Min, Thomas, Paul, Matthias and > the rest of the team. You guys done far more than your fair share of the > work, effectively picking up all of my slack in the most generous way > imaginable. It's hard to think of a better team to work with. > > Cheers, > > f > > -- > Fernando Perez (@fperez_org; http://fperez.org) > fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) > fernando.perez-at-berkeley: contact me here for any direct mail > > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > From carl.input at gmail.com Wed Nov 13 15:05:05 2013 From: carl.input at gmail.com (Carl Smith) Date: Wed, 13 Nov 2013 20:05:05 +0000 Subject: [IPython-dev] New Data Science Initiative, aka where has Fernando been hiding for the last year? In-Reply-To: <5283D8C0.3030107@telenczuk.pl> References: <5283D8C0.3030107@telenczuk.pl> Message-ID: Congratulations to everyone involved. It's awesome to see the effect your work's having on the world outside the project. Changing the way science gets done, in a way that's better for everyone, is a massive deal. All the very best for the future. -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Wed Nov 13 15:16:15 2013 From: fperez.net at gmail.com (Fernando Perez) Date: Wed, 13 Nov 2013 12:16:15 -0800 Subject: [IPython-dev] New Data Science Initiative, aka where has Fernando been hiding for the last year? In-Reply-To: <5283D8C0.3030107@telenczuk.pl> References: <5283D8C0.3030107@telenczuk.pl> Message-ID: On Wed, Nov 13, 2013 at 11:53 AM, Bartosz wrote: > Hi Fernando, > > That's great news! Congratulations! > Thanks! Crazy amount of work that was done in relatively high secrecy (which runs against my instincts, though in this case there were good reasons for it), so it's very nice to be able to talk about it publicly. > We have been organising a school on Python in (data) science (last time > in Zurich: http://python.g-node.org). The school has been a great > success, but obtaining funding for a general school without any specific > field (like bioinformatic, neuroscience etc.) is rather difficult in > Europe. > And you think it's easy here? ;) This kind of stuff is *super hard* to get funded anywhere, and it's been our struggle for over a decade. That's why we're so excited about this initiative: we hope that this support from the Moore and Sloan foundations will signal to the (much, much bigger) federal funding agencies, and the ones elsewhere in the world, that this stuff actually matters. BTW, you guys are doing a terrific job with the G-Node school, I know our own Stefan vdW is a regular and he speaks very highly of the event. Keep up the good work. Part of the reason for such a high-visibility, coordinated media blitz (White House, NSF, NYTimes, HuffPost, all universities, all foundations, blog posts, etc) was precisely to give ammunition to everyone who is fighting these battles with their committees, program directors, colleagues, etc. Hopefully next time that someone tells you this isn't important, you can point them to this collection that has all the links about this in one place: http://bitly.com/bundles/fperezorg/1 and it will make it easier to argue your case :) > Could you explain what are the main goals of the new project and what is > role Python/Ipython will play? > I really don't want to sound rude, but was there something in my blog post that wasn't quite clear about the broader picture? I'm happy to answer questions, but I'd like that post, in addition to the Berkeley press release (http://bit.ly/HYrhNT), to be a clear outline of the big ideas, so I'd rather fix that than answer only on a mailing list. As for the role for Python/IPython, right now I don't have much more to say than what was in my email: there's a *lot* of Python activity at Berkeley right now (see http://python.berkeley.edu), and multiple projects use it. Everyone in the IPython team is spending some time engaging the broader campus community, lecturing in courses, meeting students, etc. So we're spreading our engagement to campus quite a bit. What we hope is that when the institute ramps up, it will be a home for a lot of this activity; we may physically move our work space there (at least part-time), and I hope we'll be able to hire more folks from the open source Python world (and related spaces, like Julia) to make Berkeley an even more significant center of open source scientific computing. Does this help? Cheers, f -------------- next part -------------- An HTML attachment was scrubbed... URL: From mail at telenczuk.pl Wed Nov 13 21:08:31 2013 From: mail at telenczuk.pl (Bartosz) Date: Thu, 14 Nov 2013 03:08:31 +0100 Subject: [IPython-dev] =?utf-8?q?New_Data_Science_Initiative=2C_aka_where_?= =?utf-8?q?has_Fernando_been_hiding_for_the_last_year=3F?= In-Reply-To: References: <5283D8C0.3030107@telenczuk.pl> Message-ID: <1bd3878c5919f45aab0708f16b37b36e@webmail.webfaction.com> Thanks for the explanations and encouragement. Indeed, I hope that your project will open many doors for computation science. So I keep my fingers crossed. I enjoyed your blog post a lot. I definitely agree that there is need to explore the frontiers of "data science". There is much talk about it, but I wonder how much it really differs from the data analysis most scientist do. Right now I would consider it a methodology rather than science on its own. From your post I learnt about the general ideas and you did an awesome job in explaining them. I mailed to ask about specific activities planned in the project and institute (schools, online courses, software development, etc.), your reply clarified them all. Thanks! I will stay tuned. Yours, Bartosz > On Wed, Nov 13, 2013 at 11:53 AM, Bartosz wrote: > >> Hi Fernando, >> That's great news! Congratulations! > Thanks! Crazy amount of work that was done in relatively high secrecy > (which runs against my instincts, though in this case there were good > reasons for it), so it's very nice to be able to talk about it > publicly. > ? > >> We have been organising a school on Python in (data) science (last >> time >> in Zurich: http://python.g-node.org [1]). The school has been a great >> success, but obtaining funding for a general school without any >> specific >> field (like bioinformatic, neuroscience etc.) is rather difficult in >> Europe. > And you think it's easy here? ;) ?This kind of stuff is *super hard* > to get funded anywhere, and it's been our struggle for over a decade. > ?That's why we're so excited about this initiative: we hope that this > support from the Moore and Sloan foundations will signal to the (much, > much bigger) federal funding agencies, and the ones elsewhere in the > world, that this stuff actually matters. > BTW, you guys are doing a terrific job with the G-Node school, I know > our own Stefan vdW is a regular and he speaks very highly of the > event. ?Keep up the good work. > Part of the reason for such a high-visibility, coordinated media > blitz (White House, NSF, NYTimes, HuffPost, all universities, all > foundations, blog posts, etc) was precisely to give ammunition to > everyone who is fighting these battles with their committees, program > directors, colleagues, etc. ?Hopefully next time that someone tells > you this isn't important, you can point them to this collection that > has all the links about this in one place: > http://bitly.com/bundles/fperezorg/1 [2] > and it will make it easier to argue your case :) > ? > >> 39;t want to sound rude, but was there something in my blog post that >> wasn't quite clear about the broader picture > py to answer questions, but I'd like that post, in addition to the > Berkeley press release (http://bit.ly/HYrhNT [3]), to be a clear > outline of the big ideas, so I'd rather fix that than answer only on a > mailing list. > As for the role for Python/IPython, right now I don't have much more > to say than what was in my email: there's a *lot* of Python activity > at Berkeley right now (see ?http://python.berkeley.edu [4]), and > multiple projects use it. Everyone in the IPython team is spending > some time engaging the broader campus community, lecturing in courses, > meeting students, etc. So we're spreading our engagement to campus > quite a bit. What we hope is that when the institute ramps up, it will > be a home for a lot of this activity; we may physically move our work > space there (at least part-time), and I hope we'll be able to hire > more folks from the open source Python world (and related spaces, like > Julia) to make Berkeley an even more significant center of open source > scientific computing. > Does this help?? > Cheers, > f > Links: > ------ > [1] http://python.g-node.org > [2] http://bitly.com/bundles/fperezorg/1 > [3] http://bit.ly/HYrhNT > [4] http://python.berkeley.edu/ > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev From nelle.varoquaux at gmail.com Thu Nov 14 02:29:16 2013 From: nelle.varoquaux at gmail.com (Nelle Varoquaux) Date: Thu, 14 Nov 2013 08:29:16 +0100 Subject: [IPython-dev] New Data Science Initiative, aka where has Fernando been hiding for the last year? In-Reply-To: <1bd3878c5919f45aab0708f16b37b36e@webmail.webfaction.com> References: <5283D8C0.3030107@telenczuk.pl> <1bd3878c5919f45aab0708f16b37b36e@webmail.webfaction.com> Message-ID: On 14 November 2013 03:08, Bartosz wrote: > Thanks for the explanations and encouragement. Indeed, I hope that your > project will open many doors for computation science. So I keep my > fingers crossed. > > I enjoyed your blog post a lot. I definitely agree that there is need > to explore the frontiers of "data science". There is much talk about it, > but I wonder how much it really differs from the data analysis most > scientist do. Right now I would consider it a methodology rather than > science on its own. >From someone coming from "data analysis" / "computation" part of science, ouch... Luckily for me, the recent nobel price in chemistry suggests that computational sciences are beginning to be recognized as a full part of research. Data analysis is nowadays one of the most important part of fields such as biology and neuroscience because of the quantity of data generated in those fields. Sure, biologist as most scientist do "data analysis" but without understanding the underlying methods they use, and end up publishing a lot of results people cannot reproduce (something like 90% of the papers in cancer related fields cannot be reproduced). I think that developing the "methodologies" to analyse the petabytes of data we are generating is now the roadblock for big discoveries in science, in particularly in health related fields, and it is important for researchers in those fields not only to understand they need those methodologies, but also that they need *us* (in my case, the bioinformaticiens and machine learning experts). > > From your post I learnt about the general ideas and you did an awesome > job in explaining them. I mailed to ask about specific activities > planned in the project and institute (schools, online courses, software > development, etc.), your reply clarified them all. Thanks! > > I will stay tuned. > > Yours, > > Bartosz > >> On Wed, Nov 13, 2013 at 11:53 AM, Bartosz wrote: >> >>> Hi Fernando, >>> That's great news! Congratulations! >> Thanks! Crazy amount of work that was done in relatively high secrecy >> (which runs against my instincts, though in this case there were good >> reasons for it), so it's very nice to be able to talk about it >> publicly. >> >> >>> We have been organising a school on Python in (data) science (last >>> time >>> in Zurich: http://python.g-node.org [1]). The school has been a great >>> success, but obtaining funding for a general school without any >>> specific >>> field (like bioinformatic, neuroscience etc.) is rather difficult in >>> Europe. >> And you think it's easy here? ;) This kind of stuff is *super hard* >> to get funded anywhere, and it's been our struggle for over a decade. >> That's why we're so excited about this initiative: we hope that this >> support from the Moore and Sloan foundations will signal to the (much, >> much bigger) federal funding agencies, and the ones elsewhere in the >> world, that this stuff actually matters. >> BTW, you guys are doing a terrific job with the G-Node school, I know >> our own Stefan vdW is a regular and he speaks very highly of the >> event. Keep up the good work. >> Part of the reason for such a high-visibility, coordinated media >> blitz (White House, NSF, NYTimes, HuffPost, all universities, all >> foundations, blog posts, etc) was precisely to give ammunition to >> everyone who is fighting these battles with their committees, program >> directors, colleagues, etc. Hopefully next time that someone tells >> you this isn't important, you can point them to this collection that >> has all the links about this in one place: >> http://bitly.com/bundles/fperezorg/1 [2] >> and it will make it easier to argue your case :) >> >> >>> 39;t want to sound rude, but was there something in my blog post that >>> wasn't quite clear about the broader picture >> py to answer questions, but I'd like that post, in addition to the >> Berkeley press release (http://bit.ly/HYrhNT [3]), to be a clear >> outline of the big ideas, so I'd rather fix that than answer only on a >> mailing list. >> As for the role for Python/IPython, right now I don't have much more >> to say than what was in my email: there's a *lot* of Python activity >> at Berkeley right now (see http://python.berkeley.edu [4]), and >> multiple projects use it. Everyone in the IPython team is spending >> some time engaging the broader campus community, lecturing in courses, >> meeting students, etc. So we're spreading our engagement to campus >> quite a bit. What we hope is that when the institute ramps up, it will >> be a home for a lot of this activity; we may physically move our work >> space there (at least part-time), and I hope we'll be able to hire >> more folks from the open source Python world (and related spaces, like >> Julia) to make Berkeley an even more significant center of open source >> scientific computing. >> Does this help? >> Cheers, >> f >> Links: >> ------ >> [1] http://python.g-node.org >> [2] http://bitly.com/bundles/fperezorg/1 >> [3] http://bit.ly/HYrhNT >> [4] http://python.berkeley.edu/ >> _______________________________________________ >> IPython-dev mailing list >> IPython-dev at scipy.org >> http://mail.scipy.org/mailman/listinfo/ipython-dev > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev From wstein at gmail.com Thu Nov 14 02:47:47 2013 From: wstein at gmail.com (William Stein) Date: Wed, 13 Nov 2013 23:47:47 -0800 Subject: [IPython-dev] ipython with sync In-Reply-To: References: Message-ID: On Fri, Sep 13, 2013 at 10:44 AM, Brian Granger wrote: > Thanks for this link, that is helpful. I have now also BSD licensed the CoffeeScript library I wrote that implements Neil Fraser's library. It's here https://github.com/sagemath/cloud/blob/master/diffsync.coffee I'm pretty busy right now, or I would make that a nice standalone project, etc. In any case, if you want to implement sync, the above code is what I'm actually using in the client/server/etc. I hope it's useful, William > Cheers, > > Brian > > On Thu, Sep 12, 2013 at 4:04 PM, William Stein wrote: >> On Thu, Sep 12, 2013 at 2:53 PM, Brian Granger wrote: >>> William, >>> >>> Thanks for notes. This is really exciting and we are very interested >>> in getting live sync to work with IPython itself. I am traveling this >>> week, but will try to have a look soon. Is the code available >>> somewhere to look at? >> >> Unfortunately, this implementation is very entangled as part of >> something I'm building as part of a startup company with UW's Center >> for commercialization, and I can't just open source everything. But >> the algorithm is here: >> >> https://neil.fraser.name/writing/sync/ >> >> and the author of that paper has an open source Python implementation, >> I think... >> >> >>> >>> Cheers, >>> >>> Brian >>> >>> On Tue, Sep 10, 2013 at 9:11 AM, William Stein wrote: >>>> Hi, >>>> >>>> Somewhat by accident I spent the last two weeks implementing hosted >>>> IPython notebooks with sync for https://cloud.sagemath.com. >>>> Initially I had just plan to simplify the port forwarding setup, since >>>> what Ondrej Certik was doing with multiple forward and reverse port >>>> forwards seemed complicated. But then I became concerned about >>>> multiple users (or users with multiple browsers) overwriting each >>>> other's notebooks, because cloud.sagemath projects are frequently >>>> shared between multiple people, and everything else does realtime >>>> sync. I had planned just to add some very minimal merge-on-save >>>> functionality to avoid major issues, but somehow got sucked into >>>> realtime sync (even with the other person's cursor showing). >>>> >>>> It would be enormously helpful to me if a couple of expert IPython >>>> users were to try out what I implemented and just ask a bunch of >>>> questions. >>>> >>>> 1. Go to https://cloud.sagemath.com and make an account; this is >>>> completely free, and is hosted on computers at University of >>>> Washington. >>>> >>>> 2. Create a new project. >>>> >>>> 3. Click +New, then click "IPython" (or paste in a link to an ipython >>>> notebook, or upload a file). >>>> >>>> 4. An IPython notebook server will start, the given .ipynb file should >>>> load in a same-domain iframe, and then some of the ipython notebook >>>> code is and iframe contents are monkey patched, in order to support >>>> sync and better integration with https://cloud.sagemath.com. >>>> >>>> 5. Open the ipynb file in multiple browsers, and see that changes in >>>> one appear in the other, including moving cells around, creating new >>>> cells, editing markdown (the rendered version appears elsewhere), etc. >>>> Anything that sets the notebook.dirty flag in IPython causes a sync >>>> (evaluating a cell that creates no output doesn't set this flag, at >>>> least in 1.0.0, which is a bug in IPython, I guess). >>>> >>>> Since this is all very new and the first (I guess) realtime sync >>>> implementation on top of IPython, there are probably a lot of issues. >>>> Note that if you click the "i" info button to the right, you'll get a >>>> link to the standard IPython >>>> >>>> The other thing of interest is a little Python script called >>>> "ipython-notebook", which I wrote. It basically makes it easy to run >>>> an IPython notebook server as a daemon, get the port it is running on, >>>> etc. It's pretty simple but satisfies my use case, and has >>>> JSON-output, to make it web friendly. As I've written it, my script >>>> passes several base_url options through by default, which are needed >>>> for cloud.sagemath. Anyway, I've attached it to this email (with a >>>> BSD license) in case there is any interest. >>>> >>>> Regarding the monkey patching in 4 above, the right thing to do would >>>> be to explain exactly what hooks/changes in the IPython html client I >>>> need in order to do sync, etc., make sure these makes sense to the >>>> IPython devs, and send a pull request (or have a coding sprint in >>>> Seattle or Berkeley?). As an example, in order to do sync >>>> *efficiently*, I have to be able to set a given cell from JSON -- it's >>>> critical to do this in place when possible, since the overhead of >>>> creating a new cell is huge (due probably to the overhead of creating >>>> CodeMirror editors); however, the fromJSON method in IPython assumes >>>> that the cell is brand new -- it would be nice to add an option to >>>> make a cell fromJSON without assuming it is empty. >>>> >>>> The ultimate outcome of this could be a clean well-defined way of >>>> doing sync for IPython notebooks using any third-party sync >>>> implementation. IPython might provide their own sync service and >>>> there are starting to be others available these days -- e.g., Google >>>> has one: https://developers.google.com/drive/realtime/, and maybe >>>> Guido van Rosum helped write one for Dropbox recently? >>>> >>>> Subdirectories: I noticed, incidentally, that the wakari version of >>>> the IPython notebook server allows one to load ipynb files that are in >>>> any subdirectory, whereas the standard IPython notebook server >>>> doesn't. For cloud.sagemath, I just spawn a new IPython notebook >>>> server for each directory that a user accesses files in right now. >>>> This seems cludgy, so I'm interested in the situation regarding adding >>>> support for subdirectories. >>>> >>>> -- William >>>> >>>> >>>> >>>> -- >>>> William Stein >>>> Professor of Mathematics >>>> University of Washington >>>> http://wstein.org >>>> >>>> _______________________________________________ >>>> IPython-dev mailing list >>>> IPython-dev at scipy.org >>>> http://mail.scipy.org/mailman/listinfo/ipython-dev >>>> >>> >>> >>> >>> -- >>> Brian E. Granger >>> Cal Poly State University, San Luis Obispo >>> bgranger at calpoly.edu and ellisonbg at gmail.com >>> _______________________________________________ >>> IPython-dev mailing list >>> IPython-dev at scipy.org >>> http://mail.scipy.org/mailman/listinfo/ipython-dev >> >> >> >> -- >> William Stein >> Professor of Mathematics >> University of Washington >> http://wstein.org >> _______________________________________________ >> IPython-dev mailing list >> IPython-dev at scipy.org >> http://mail.scipy.org/mailman/listinfo/ipython-dev > > > > -- > Brian E. Granger > Cal Poly State University, San Luis Obispo > bgranger at calpoly.edu and ellisonbg at gmail.com > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev -- William Stein Professor of Mathematics University of Washington http://wstein.org From ozancag at gmail.com Thu Nov 14 09:16:50 2013 From: ozancag at gmail.com (=?UTF-8?B?T3phbiDDh2HEn2xheWFu?=) Date: Thu, 14 Nov 2013 16:16:50 +0200 Subject: [IPython-dev] New Data Science Initiative, aka where has Fernando been hiding for the last year? In-Reply-To: References: Message-ID: On Wed, Nov 13, 2013 at 8:23 AM, Fernando Perez wrote: > Hi all, > This became a very significant and time consuming project, but today we've > been able to publicly announce the outcome, during an event at the White > House OSTP. Rather than repeating in this email everything, I'll just point > to a blog post I wrote with the rest of the story: > > http://blog.fperez.org/2013/11/an-ambitious-experiment-in-data-science.html Hi Fernando, That is such a big encouragement to all of us. As a research assistant from Turkey, I was quite depressed from how things are working in the academy, e.g. papers, citations, lead authors, attributions, no code, no replication, etc. But I read your blog post and got some pointers from that and read a lot of other stuff and now I know that I am not alone (at least throughout the world) But, I am quite alone in Turkey in these terms, so I just bought a domain acikbilim.net (which means openscience.net) to spread news about reproducible science, sharing codes, making people aware of the open science tools, Python ecosystem, etc. I'd like to translate blog posts and articles about the subject to Turkish and publish through there. This community will change/is changing the world and you just motivated me to do this in a Turkey. Many thanks. -- Ozan ?a?layan Research Assistant Galatasaray University - Computer Engineering Dept. http://www.ozancaglayan.com From ellisonbg at gmail.com Thu Nov 14 15:27:26 2013 From: ellisonbg at gmail.com (Brian Granger) Date: Thu, 14 Nov 2013 15:27:26 -0500 Subject: [IPython-dev] ipython with sync In-Reply-To: References: Message-ID: Thanks William, will have a look at it! Cheers, Brian On Thu, Nov 14, 2013 at 2:47 AM, William Stein wrote: > On Fri, Sep 13, 2013 at 10:44 AM, Brian Granger wrote: >> Thanks for this link, that is helpful. > > I have now also BSD licensed the CoffeeScript library I wrote that > implements Neil Fraser's library. It's here > > https://github.com/sagemath/cloud/blob/master/diffsync.coffee > > I'm pretty busy right now, or I would make that a nice standalone > project, etc. In any case, if you want to implement sync, the above > code is what I'm actually using in the client/server/etc. > > I hope it's useful, > > William > > >> Cheers, >> >> Brian >> >> On Thu, Sep 12, 2013 at 4:04 PM, William Stein wrote: >>> On Thu, Sep 12, 2013 at 2:53 PM, Brian Granger wrote: >>>> William, >>>> >>>> Thanks for notes. This is really exciting and we are very interested >>>> in getting live sync to work with IPython itself. I am traveling this >>>> week, but will try to have a look soon. Is the code available >>>> somewhere to look at? >>> >>> Unfortunately, this implementation is very entangled as part of >>> something I'm building as part of a startup company with UW's Center >>> for commercialization, and I can't just open source everything. But >>> the algorithm is here: >>> >>> https://neil.fraser.name/writing/sync/ >>> >>> and the author of that paper has an open source Python implementation, >>> I think... >>> >>> >>>> >>>> Cheers, >>>> >>>> Brian >>>> >>>> On Tue, Sep 10, 2013 at 9:11 AM, William Stein wrote: >>>>> Hi, >>>>> >>>>> Somewhat by accident I spent the last two weeks implementing hosted >>>>> IPython notebooks with sync for https://cloud.sagemath.com. >>>>> Initially I had just plan to simplify the port forwarding setup, since >>>>> what Ondrej Certik was doing with multiple forward and reverse port >>>>> forwards seemed complicated. But then I became concerned about >>>>> multiple users (or users with multiple browsers) overwriting each >>>>> other's notebooks, because cloud.sagemath projects are frequently >>>>> shared between multiple people, and everything else does realtime >>>>> sync. I had planned just to add some very minimal merge-on-save >>>>> functionality to avoid major issues, but somehow got sucked into >>>>> realtime sync (even with the other person's cursor showing). >>>>> >>>>> It would be enormously helpful to me if a couple of expert IPython >>>>> users were to try out what I implemented and just ask a bunch of >>>>> questions. >>>>> >>>>> 1. Go to https://cloud.sagemath.com and make an account; this is >>>>> completely free, and is hosted on computers at University of >>>>> Washington. >>>>> >>>>> 2. Create a new project. >>>>> >>>>> 3. Click +New, then click "IPython" (or paste in a link to an ipython >>>>> notebook, or upload a file). >>>>> >>>>> 4. An IPython notebook server will start, the given .ipynb file should >>>>> load in a same-domain iframe, and then some of the ipython notebook >>>>> code is and iframe contents are monkey patched, in order to support >>>>> sync and better integration with https://cloud.sagemath.com. >>>>> >>>>> 5. Open the ipynb file in multiple browsers, and see that changes in >>>>> one appear in the other, including moving cells around, creating new >>>>> cells, editing markdown (the rendered version appears elsewhere), etc. >>>>> Anything that sets the notebook.dirty flag in IPython causes a sync >>>>> (evaluating a cell that creates no output doesn't set this flag, at >>>>> least in 1.0.0, which is a bug in IPython, I guess). >>>>> >>>>> Since this is all very new and the first (I guess) realtime sync >>>>> implementation on top of IPython, there are probably a lot of issues. >>>>> Note that if you click the "i" info button to the right, you'll get a >>>>> link to the standard IPython >>>>> >>>>> The other thing of interest is a little Python script called >>>>> "ipython-notebook", which I wrote. It basically makes it easy to run >>>>> an IPython notebook server as a daemon, get the port it is running on, >>>>> etc. It's pretty simple but satisfies my use case, and has >>>>> JSON-output, to make it web friendly. As I've written it, my script >>>>> passes several base_url options through by default, which are needed >>>>> for cloud.sagemath. Anyway, I've attached it to this email (with a >>>>> BSD license) in case there is any interest. >>>>> >>>>> Regarding the monkey patching in 4 above, the right thing to do would >>>>> be to explain exactly what hooks/changes in the IPython html client I >>>>> need in order to do sync, etc., make sure these makes sense to the >>>>> IPython devs, and send a pull request (or have a coding sprint in >>>>> Seattle or Berkeley?). As an example, in order to do sync >>>>> *efficiently*, I have to be able to set a given cell from JSON -- it's >>>>> critical to do this in place when possible, since the overhead of >>>>> creating a new cell is huge (due probably to the overhead of creating >>>>> CodeMirror editors); however, the fromJSON method in IPython assumes >>>>> that the cell is brand new -- it would be nice to add an option to >>>>> make a cell fromJSON without assuming it is empty. >>>>> >>>>> The ultimate outcome of this could be a clean well-defined way of >>>>> doing sync for IPython notebooks using any third-party sync >>>>> implementation. IPython might provide their own sync service and >>>>> there are starting to be others available these days -- e.g., Google >>>>> has one: https://developers.google.com/drive/realtime/, and maybe >>>>> Guido van Rosum helped write one for Dropbox recently? >>>>> >>>>> Subdirectories: I noticed, incidentally, that the wakari version of >>>>> the IPython notebook server allows one to load ipynb files that are in >>>>> any subdirectory, whereas the standard IPython notebook server >>>>> doesn't. For cloud.sagemath, I just spawn a new IPython notebook >>>>> server for each directory that a user accesses files in right now. >>>>> This seems cludgy, so I'm interested in the situation regarding adding >>>>> support for subdirectories. >>>>> >>>>> -- William >>>>> >>>>> >>>>> >>>>> -- >>>>> William Stein >>>>> Professor of Mathematics >>>>> University of Washington >>>>> http://wstein.org >>>>> >>>>> _______________________________________________ >>>>> IPython-dev mailing list >>>>> IPython-dev at scipy.org >>>>> http://mail.scipy.org/mailman/listinfo/ipython-dev >>>>> >>>> >>>> >>>> >>>> -- >>>> Brian E. Granger >>>> Cal Poly State University, San Luis Obispo >>>> bgranger at calpoly.edu and ellisonbg at gmail.com >>>> _______________________________________________ >>>> IPython-dev mailing list >>>> IPython-dev at scipy.org >>>> http://mail.scipy.org/mailman/listinfo/ipython-dev >>> >>> >>> >>> -- >>> William Stein >>> Professor of Mathematics >>> University of Washington >>> http://wstein.org >>> _______________________________________________ >>> IPython-dev mailing list >>> IPython-dev at scipy.org >>> http://mail.scipy.org/mailman/listinfo/ipython-dev >> >> >> >> -- >> Brian E. Granger >> Cal Poly State University, San Luis Obispo >> bgranger at calpoly.edu and ellisonbg at gmail.com >> _______________________________________________ >> IPython-dev mailing list >> IPython-dev at scipy.org >> http://mail.scipy.org/mailman/listinfo/ipython-dev > > > > -- > William Stein > Professor of Mathematics > University of Washington > http://wstein.org > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev -- Brian E. Granger Cal Poly State University, San Luis Obispo bgranger at calpoly.edu and ellisonbg at gmail.com From asadovsky at gmail.com Thu Nov 14 15:32:09 2013 From: asadovsky at gmail.com (Adam Sadovsky) Date: Thu, 14 Nov 2013 12:32:09 -0800 Subject: [IPython-dev] How to preconfigure notebook environment in v1.1? In-Reply-To: References: <20131031041109.GS12247@HbI-OTOH.berkeley.edu> Message-ID: For the record, I ended up achieving this by modifying KernelManager to invoke the original script in a subprocess (rather than just calling zmq main), and then configuring that binary, when called in kernel mode (i.e. with "kernel" argument), to call IPython.start_kernel with the desired user_ns passed in. Is there any documentation for developers who want to better understand the internal IPython APIs, specifically around kernels, applications, etc.? I felt myself digging around somewhat blindly in order to figure this out, and I'm sure there are better ways to do it (that don't require modifying IPython internals). --Adam On Wed, Oct 30, 2013 at 10:32 PM, Adam Sadovsky wrote: > Hi Paul, > > Thanks for the reply! The reason I'd like to avoid startup files is that > I'm trying to provide a library for others to use. As part of this library, > I want it to be possible for people to start a notebook server where any > new notebook has a certain set of things pre-imported for convenience. I > suppose I could provide a startup file for users to place in their own > startup dirs, but I'd like to avoid making users do any extra work. Ideally > I'd like an option similar to what's offered in the interactive shell API, > which allows users to pass in a user_ns object. Does that make sense? > > I should say, I'm relatively new to IPython/notebook and not very familiar > with its architecture, so I certainly may be missing something obvious. :) > > > --Adam > > > On Wed, Oct 30, 2013 at 9:11 PM, Paul Ivanov wrote: > >> Adam Sadovsky, on 2013-10-30 20:28, wrote: >> > I recently upgraded from IPython 0.13 to 1.1. >> > In 0.13, I was able to execute code in each notebook at startup, but the >> > kernel code has changed and I can't figure out how to do it anymore. >> Could >> > someone point me in the right direction? >> >> Hi Adam, >> >> why not just use a file in the startup directory? >> >> $ echo `ipython locate profile`/startup/ >> /home/pi/.ipython/profile_default/startup/ >> $ cat `ipython locate profile`/startup/README >> This is the IPython startup directory >> >> .py and .ipy files in this directory will be run *prior* to any >> code or files specified >> via the exec_lines or exec_files configurables whenever you load >> this profile. >> >> Files will be run in lexicographical order, so you can control >> the execution order of files >> with a prefix, e.g.:: >> >> 00-first.py >> 50-middle.py >> 99-last.ipy >> >> best, >> -- >> _ >> / \ >> A* \^ - >> ,./ _.`\\ / \ >> / ,--.S \/ \ >> / `"~,_ \ \ >> __o ? >> _ \<,_ /:\ >> --(_)/-(_)----.../ | \ >> --------------.......J >> Paul Ivanov >> http://pirsquared.org >> _______________________________________________ >> IPython-dev mailing list >> IPython-dev at scipy.org >> http://mail.scipy.org/mailman/listinfo/ipython-dev >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From bussonniermatthias at gmail.com Fri Nov 15 03:44:28 2013 From: bussonniermatthias at gmail.com (Matthias BUSSONNIER) Date: Fri, 15 Nov 2013 09:44:28 +0100 Subject: [IPython-dev] [IPython-User] New Data Science Initiative, aka where has Fernando been hiding for the last year? In-Reply-To: References: Message-ID: Le 14 nov. 2013 ? 22:41, Brian Granger a ?crit : > David, >> >> Just wanted to let you know that IPython is much appreciated in the VFX >> community of New York, of which I am a part. So when you see some of the >> commercials on TV, during the Super Bowl or otherwise, IPython may very well >> be playing a role in there. >> >> This current effort is a very encouraging development for the potential to >> contribute back to open source. It would be great to have a legitimate means >> to bring some of the lessons and strategies of production VFX work to other >> areas. > > Thanks for sharing about how the VFX community is using IPython! It is > fantastic to find out about usage cases like this that are further > away from scientific computing. Agree this is great, speaking of VFX, has anyone news about integration of IPython and Blender ? I remember an old thread about that somewhere. David, Could you enlighten us a little on how you use IPython in VFX ? (would love to see a before/after IPython picture overlay :-D ) This make me think that it could be a good idea to gather testimony (correct term?) of how people use IPython on the website. What do others think ? -- Matthias Note, re-including ipyton-dev list. -------------- next part -------------- An HTML attachment was scrubbed... URL: From p.f.moore at gmail.com Fri Nov 15 04:26:21 2013 From: p.f.moore at gmail.com (Paul Moore) Date: Fri, 15 Nov 2013 09:26:21 +0000 Subject: [IPython-dev] Dependency on pyreadline In-Reply-To: References: Message-ID: On 13 November 2013 08:14, Paul Moore wrote: >> Do you want to do a quick PR against the docs? > > I'd be happy to, may take a couple of days to find some free time, but > I'll sort something out. I'm just looking at this now. One thing that struck me was that the reason I picked up on this is that when installed, pyreadline in injected into *every* Python process (because core python imports readline on startup). This changes behaviour in *every* Python process (for example, "standard" Windows command history is lost, and Ctrl-Z works differently). Also, there is a bug of some sort (involving the low level console handling) that causes an infinite exception chain if you import pip from the interactive prompt. Is it appropriate to mention these issues in the IPython docs? I don't want to add FUD, but I think people should have the information they need to make a decision. (Maybe add the point about changed behaviour, but omit the bug as it's rare, and might get fixed). Paul From jason-sage at creativetrax.com Fri Nov 15 09:07:39 2013 From: jason-sage at creativetrax.com (Jason Grout) Date: Fri, 15 Nov 2013 08:07:39 -0600 Subject: [IPython-dev] interactive matplotlib via comm infrastructure Message-ID: <52862AAB.70206@creativetrax.com> I should probably have cross-posted this: http://sourceforge.net/mailarchive/message.php?msg_id=31640342 Following a very helpful conversation with Michael yesterday in the dev hangout, I got the interactive matplotlib working with the current master (of matplotlib and ipython). I updated the pull request at https://github.com/matplotlib/matplotlib/pull/2524 To test this, run IPython (master branch, to get the comm commits), and put this in a cell: https://github.com/matplotlib/matplotlib/pull/2524#issuecomment-28539813 Then you can execute something like: from matplotlib.figure import Figure import numpy as np fig = Figure() a = fig.add_subplot(111) t = np.arange(0.0, 3.0, 0.01) s = np.sin(2 * np.pi * t) a.plot(t, s) CommFigure(fig) and get a live figure in the IPython notebook that uses the comm messaging infrastructure. I also got this working in the Sage cell server: http://sagecell.sagemath.org/?q=ilpajg (for a sage plotting example) http://sagecell.sagemath.org/?q=spadja (for one of the examples from the matplotlib docs) Thanks again for everyone's work on the comm infrastructure! Jason From p.f.moore at gmail.com Fri Nov 15 09:43:27 2013 From: p.f.moore at gmail.com (Paul Moore) Date: Fri, 15 Nov 2013 14:43:27 +0000 Subject: [IPython-dev] Dependency on pyreadline In-Reply-To: References: Message-ID: On 13 November 2013 08:14, Paul Moore wrote: >> Do you want to do a quick PR against the docs? > > I'd be happy to, may take a couple of days to find some free time, but > I'll sort something out. https://github.com/ipython/ipython/pull/4544 From takowl at gmail.com Fri Nov 15 20:24:20 2013 From: takowl at gmail.com (Thomas Kluyver) Date: Fri, 15 Nov 2013 17:24:20 -0800 Subject: [IPython-dev] Go IPython kernel Message-ID: I've been working on a simple kernel for the Go language. My main reason for writing this was to learn Go, so forgive my awkward code. The basics are now working, and you can get it from here: https://github.com/takluyver/igo At the moment, this is something of a proof of concept. Go is a compiled language, and it doesn't support dynamic execution natively. I'm using a third-party package (https://github.com/sbinet/go-eval ) for execution, but it doesn't support some key things (like imports - see bug 1). So if anyone wants to use this seriously, you should look at improving that. Thanks, Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From ehmatthes at gmail.com Mon Nov 18 19:34:26 2013 From: ehmatthes at gmail.com (Eric Matthes) Date: Mon, 18 Nov 2013 15:34:26 -0900 Subject: [IPython-dev] Accessing python 2 and python 3 from the same virtualenv Message-ID: Hi, I am working on a set of notebooks that form the basis of a Python curriculum, so sometimes I want to use a Python 3 interpreter and sometimes I want to use a Python 2 interpreter. I am running Ubuntu 12.04, with a number of different Python versions installed. To access any particular version, I just enter `python` in a terminal for the default Python, `python2.7` for a 2.7 interpreter, `python3.2`, etc. When I set up a virtualenv and install ipython, ipython attaches to a particular python. What I'd like to do: - Set up a virtualenv. - Install ipython. - Be able to run `ipython notebook` attached to Python 2.7 - Be able to run `ipython notebook` attached to Python 3.2 Is there a straightforward way to do this? I don't want to have to maintain two separate virtualenv's, if possible. Thank you. Eric -------------- next part -------------- An HTML attachment was scrubbed... URL: From takowl at gmail.com Mon Nov 18 19:47:30 2013 From: takowl at gmail.com (Thomas Kluyver) Date: Mon, 18 Nov 2013 16:47:30 -0800 Subject: [IPython-dev] Accessing python 2 and python 3 from the same virtualenv In-Reply-To: References: Message-ID: Hi Eric, On 18 November 2013 16:34, Eric Matthes wrote: > Is there a straightforward way to do this? I don't want to have to > maintain two separate virtualenv's, if possible. > Not really. A virtualenv is associated with a specific version of Python, so if you want to use two versions of Python, you'll need two separate virtualenvs. If you want to avoid the hassle of virtualenvs, you could just use the system Python, and install packages with the --user flag. Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From haoxing.zhang.david at gmail.com Tue Nov 19 17:44:21 2013 From: haoxing.zhang.david at gmail.com (Haoxing Zhang) Date: Tue, 19 Nov 2013 14:44:21 -0800 Subject: [IPython-dev] Hi from new developer Message-ID: Hi everyone, My name is Haoxing Zhang. I am a graduate student in Stanford University. I recently started working with Ariel Rokem on the python-matlab-bridge project. I feel very excited to join this developer community and I believe I can learn a lot here. The following is the github link of our project: https://github.com/arokem/python-matlab-bridge Best, Haoxing -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Wed Nov 20 00:18:24 2013 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 19 Nov 2013 21:18:24 -0800 Subject: [IPython-dev] Hi from new developer In-Reply-To: References: Message-ID: Hi Haoxing, welcome to the list! We look forward to progress on the matlab integration, don't hesitate to ask any questions you may have. Cheers, f On Tue, Nov 19, 2013 at 2:44 PM, Haoxing Zhang < haoxing.zhang.david at gmail.com> wrote: > Hi everyone, > > My name is Haoxing Zhang. I am a graduate student in Stanford University. > I recently started working with Ariel Rokem on the python-matlab-bridge > project. I feel very excited to join this developer community and I believe > I can learn a lot here. > > The following is the github link of our project: > https://github.com/arokem/python-matlab-bridge > > Best, > Haoxing > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > > -- Fernando Perez (@fperez_org; http://fperez.org) fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) fernando.perez-at-berkeley: contact me here for any direct mail -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Wed Nov 20 01:31:49 2013 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 19 Nov 2013 22:31:49 -0800 Subject: [IPython-dev] [IPython-User] New Data Science Initiative, aka where has Fernando been hiding for the last year? In-Reply-To: References: Message-ID: On Wed, Nov 13, 2013 at 7:30 AM, David Cournapeau wrote: > Congrats to everybody involved, that's very exciting ! > Thanks for the kind words :) And I really want to reiterate: I completely see this as a victory for our entire community at large. 12 years of hard work (and counting) on scientific python are finally starting to pay off in significant ways. > > Are we still able to speak to you face to face now that you went to the > white house ? :) > Well, I wasn't able to make it, so I had to send one of our Nobel laureates in my stead ;) Cheers, f -- Fernando Perez (@fperez_org; http://fperez.org) fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) fernando.perez-at-berkeley: contact me here for any direct mail -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Wed Nov 20 01:33:06 2013 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 19 Nov 2013 22:33:06 -0800 Subject: [IPython-dev] interactive matplotlib via comm infrastructure In-Reply-To: <52862AAB.70206@creativetrax.com> References: <52862AAB.70206@creativetrax.com> Message-ID: Hi Jason, this is great news! Very much looking forward to seeing this work its way into master for both projects, and eventually likely to become the standard. Very happy to see that all the time we put into the design of the comm machinery is paying off. Cheers, f On Fri, Nov 15, 2013 at 6:07 AM, Jason Grout wrote: > I should probably have cross-posted this: > http://sourceforge.net/mailarchive/message.php?msg_id=31640342 > > Following a very helpful conversation with Michael yesterday in the > dev hangout, I got the interactive matplotlib working with the current > master (of matplotlib and ipython). > > I updated the pull request at > https://github.com/matplotlib/matplotlib/pull/2524 > > To test this, run IPython (master branch, to get the comm commits), and > put this in a cell: > https://github.com/matplotlib/matplotlib/pull/2524#issuecomment-28539813 > > Then you can execute something like: > > from matplotlib.figure import Figure > import numpy as np > fig = Figure() > a = fig.add_subplot(111) > t = np.arange(0.0, 3.0, 0.01) > s = np.sin(2 * np.pi * t) > a.plot(t, s) > CommFigure(fig) > > and get a live figure in the IPython notebook that uses the comm > messaging infrastructure. > > I also got this working in the Sage cell server: > > http://sagecell.sagemath.org/?q=ilpajg (for a sage plotting example) > > http://sagecell.sagemath.org/?q=spadja (for one of the examples from the > matplotlib docs) > > Thanks again for everyone's work on the comm infrastructure! > > Jason > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > -- Fernando Perez (@fperez_org; http://fperez.org) fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) fernando.perez-at-berkeley: contact me here for any direct mail -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Wed Nov 20 01:33:43 2013 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 19 Nov 2013 22:33:43 -0800 Subject: [IPython-dev] ipython with sync In-Reply-To: References: Message-ID: Awesome, William; many thanks! Very cool on your part. Cheers, f On Wed, Nov 13, 2013 at 11:47 PM, William Stein wrote: > On Fri, Sep 13, 2013 at 10:44 AM, Brian Granger > wrote: > > Thanks for this link, that is helpful. > > I have now also BSD licensed the CoffeeScript library I wrote that > implements Neil Fraser's library. It's here > > https://github.com/sagemath/cloud/blob/master/diffsync.coffee > > I'm pretty busy right now, or I would make that a nice standalone > project, etc. In any case, if you want to implement sync, the above > code is what I'm actually using in the client/server/etc. > > I hope it's useful, > > William > > > > Cheers, > > > > Brian > > > > On Thu, Sep 12, 2013 at 4:04 PM, William Stein wrote: > >> On Thu, Sep 12, 2013 at 2:53 PM, Brian Granger > wrote: > >>> William, > >>> > >>> Thanks for notes. This is really exciting and we are very interested > >>> in getting live sync to work with IPython itself. I am traveling this > >>> week, but will try to have a look soon. Is the code available > >>> somewhere to look at? > >> > >> Unfortunately, this implementation is very entangled as part of > >> something I'm building as part of a startup company with UW's Center > >> for commercialization, and I can't just open source everything. But > >> the algorithm is here: > >> > >> https://neil.fraser.name/writing/sync/ > >> > >> and the author of that paper has an open source Python implementation, > >> I think... > >> > >> > >>> > >>> Cheers, > >>> > >>> Brian > >>> > >>> On Tue, Sep 10, 2013 at 9:11 AM, William Stein > wrote: > >>>> Hi, > >>>> > >>>> Somewhat by accident I spent the last two weeks implementing hosted > >>>> IPython notebooks with sync for https://cloud.sagemath.com. > >>>> Initially I had just plan to simplify the port forwarding setup, since > >>>> what Ondrej Certik was doing with multiple forward and reverse port > >>>> forwards seemed complicated. But then I became concerned about > >>>> multiple users (or users with multiple browsers) overwriting each > >>>> other's notebooks, because cloud.sagemath projects are frequently > >>>> shared between multiple people, and everything else does realtime > >>>> sync. I had planned just to add some very minimal merge-on-save > >>>> functionality to avoid major issues, but somehow got sucked into > >>>> realtime sync (even with the other person's cursor showing). > >>>> > >>>> It would be enormously helpful to me if a couple of expert IPython > >>>> users were to try out what I implemented and just ask a bunch of > >>>> questions. > >>>> > >>>> 1. Go to https://cloud.sagemath.com and make an account; this is > >>>> completely free, and is hosted on computers at University of > >>>> Washington. > >>>> > >>>> 2. Create a new project. > >>>> > >>>> 3. Click +New, then click "IPython" (or paste in a link to an ipython > >>>> notebook, or upload a file). > >>>> > >>>> 4. An IPython notebook server will start, the given .ipynb file should > >>>> load in a same-domain iframe, and then some of the ipython notebook > >>>> code is and iframe contents are monkey patched, in order to support > >>>> sync and better integration with https://cloud.sagemath.com. > >>>> > >>>> 5. Open the ipynb file in multiple browsers, and see that changes in > >>>> one appear in the other, including moving cells around, creating new > >>>> cells, editing markdown (the rendered version appears elsewhere), etc. > >>>> Anything that sets the notebook.dirty flag in IPython causes a sync > >>>> (evaluating a cell that creates no output doesn't set this flag, at > >>>> least in 1.0.0, which is a bug in IPython, I guess). > >>>> > >>>> Since this is all very new and the first (I guess) realtime sync > >>>> implementation on top of IPython, there are probably a lot of issues. > >>>> Note that if you click the "i" info button to the right, you'll get a > >>>> link to the standard IPython > >>>> > >>>> The other thing of interest is a little Python script called > >>>> "ipython-notebook", which I wrote. It basically makes it easy to run > >>>> an IPython notebook server as a daemon, get the port it is running on, > >>>> etc. It's pretty simple but satisfies my use case, and has > >>>> JSON-output, to make it web friendly. As I've written it, my script > >>>> passes several base_url options through by default, which are needed > >>>> for cloud.sagemath. Anyway, I've attached it to this email (with a > >>>> BSD license) in case there is any interest. > >>>> > >>>> Regarding the monkey patching in 4 above, the right thing to do would > >>>> be to explain exactly what hooks/changes in the IPython html client I > >>>> need in order to do sync, etc., make sure these makes sense to the > >>>> IPython devs, and send a pull request (or have a coding sprint in > >>>> Seattle or Berkeley?). As an example, in order to do sync > >>>> *efficiently*, I have to be able to set a given cell from JSON -- it's > >>>> critical to do this in place when possible, since the overhead of > >>>> creating a new cell is huge (due probably to the overhead of creating > >>>> CodeMirror editors); however, the fromJSON method in IPython assumes > >>>> that the cell is brand new -- it would be nice to add an option to > >>>> make a cell fromJSON without assuming it is empty. > >>>> > >>>> The ultimate outcome of this could be a clean well-defined way of > >>>> doing sync for IPython notebooks using any third-party sync > >>>> implementation. IPython might provide their own sync service and > >>>> there are starting to be others available these days -- e.g., Google > >>>> has one: https://developers.google.com/drive/realtime/, and maybe > >>>> Guido van Rosum helped write one for Dropbox recently? > >>>> > >>>> Subdirectories: I noticed, incidentally, that the wakari version of > >>>> the IPython notebook server allows one to load ipynb files that are in > >>>> any subdirectory, whereas the standard IPython notebook server > >>>> doesn't. For cloud.sagemath, I just spawn a new IPython notebook > >>>> server for each directory that a user accesses files in right now. > >>>> This seems cludgy, so I'm interested in the situation regarding adding > >>>> support for subdirectories. > >>>> > >>>> -- William > >>>> > >>>> > >>>> > >>>> -- > >>>> William Stein > >>>> Professor of Mathematics > >>>> University of Washington > >>>> http://wstein.org > >>>> > >>>> _______________________________________________ > >>>> IPython-dev mailing list > >>>> IPython-dev at scipy.org > >>>> http://mail.scipy.org/mailman/listinfo/ipython-dev > >>>> > >>> > >>> > >>> > >>> -- > >>> Brian E. Granger > >>> Cal Poly State University, San Luis Obispo > >>> bgranger at calpoly.edu and ellisonbg at gmail.com > >>> _______________________________________________ > >>> IPython-dev mailing list > >>> IPython-dev at scipy.org > >>> http://mail.scipy.org/mailman/listinfo/ipython-dev > >> > >> > >> > >> -- > >> William Stein > >> Professor of Mathematics > >> University of Washington > >> http://wstein.org > >> _______________________________________________ > >> IPython-dev mailing list > >> IPython-dev at scipy.org > >> http://mail.scipy.org/mailman/listinfo/ipython-dev > > > > > > > > -- > > Brian E. Granger > > Cal Poly State University, San Luis Obispo > > bgranger at calpoly.edu and ellisonbg at gmail.com > > _______________________________________________ > > IPython-dev mailing list > > IPython-dev at scipy.org > > http://mail.scipy.org/mailman/listinfo/ipython-dev > > > > -- > William Stein > Professor of Mathematics > University of Washington > http://wstein.org > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > -- Fernando Perez (@fperez_org; http://fperez.org) fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) fernando.perez-at-berkeley: contact me here for any direct mail -------------- next part -------------- An HTML attachment was scrubbed... URL: From jorgen.stenarson at kroywen.se Wed Nov 20 03:25:40 2013 From: jorgen.stenarson at kroywen.se (J=?iso-8859-1?Q?=F6?=rgen Stenarson) Date: Wed, 20 Nov 2013 08:25:40 GMT Subject: [IPython-dev] Dependency on pyreadline In-Reply-To: Message-ID: <1384935940534.31207.4220@webmail7> Strictly speaking it is not imported into *every* python process. Only those that are interactive, for a script invoked by python script.py, pyreadline is not imported unless you import for instance pdb and use set_trace. If you add?disable_readline(True) to the pyreadline.ini file pyreadline should not integrate itself into the readline hooks of python and you would see the standard behaviour where ctrl-D will generate ^D at the prompt instead of exiting the process. I have not seen the infinite exception loop you see when using pyreadline master (what version of pyreadline and python are you using?) C:\python27> python Python 2.7.3 (default, Apr 10 2012, 23:31:26) [MSC v.1500 32 bit (Intel)] on win32 Type "help", "copyright", "credits" or "license" for more information. >>> import pip >>> Fri Nov 15 2013 10:26:21 GMT+0100 (V?steuropa, normaltid), Paul Moore skrev: > On 13 November 2013 08:14, Paul Moore <> wrote: > > > > Do you want to do a quick PR against the docs? > > > > > > > I'd be happy to, may take a couple of days to find some free time, but > > I'll sort something out. > > > > I'm just looking at this now. One thing that struck me was that the > reason I picked up on this is that when installed, pyreadline in > injected into *every* Python process (because core python imports > readline on startup). This changes behaviour in *every* Python process > (for example, "standard" Windows command history is lost, and Ctrl-Z > works differently). Also, there is a bug of some sort (involving the > low level console handling) that causes an infinite exception chain if > you import pip from the interactive prompt. > > Is it appropriate to mention these issues in the IPython docs? I don't > want to add FUD, but I think people should have the information they > need to make a decision. (Maybe add the point about changed behaviour, > but omit the bug as it's rare, and might get fixed). > > Paul > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From p.f.moore at gmail.com Wed Nov 20 04:21:07 2013 From: p.f.moore at gmail.com (Paul Moore) Date: Wed, 20 Nov 2013 09:21:07 +0000 Subject: [IPython-dev] Dependency on pyreadline In-Reply-To: <1384935940534.31207.4220@webmail7> References: <1384935940534.31207.4220@webmail7> Message-ID: On 20 November 2013 08:25, J?rgen Stenarson wrote: > Strictly speaking it is not imported into *every* python process. Only those > that are interactive, for a script invoked by python script.py, pyreadline > is not imported unless you import for instance pdb and use set_trace. Did I not say interactive? Sorry, yes it's only interactive processes. > If you add disable_readline(True) to the pyreadline.ini file pyreadline > should not integrate itself into the readline hooks of python and you would > see the standard behaviour where ctrl-D will generate ^D at the prompt > instead of exiting the process. Hmm, I wasn't even aware of pyreadline.ini - where is that documented? Regardless, that is presumably a global file, so how is adding a line to that file different from uninstalling pyreadline? (Both are easy enough to do in a virtualenv, just inconvenient) > I have not seen the infinite exception loop you see when using pyreadline > master (what version of pyreadline and python are you using?) > > C:\python27> python > Python 2.7.3 (default, Apr 10 2012, 23:31:26) [MSC v.1500 32 bit (Intel)] on > win32 > Type "help", "copyright", "credits" or "license" for more information. >>>> import pip >>>> Python 3.3.2, whatever the latest released version is of Pyreadline (2.0, see below), and pip development version (1.5pre) from git (pip dev is important, see below). >virtualenv foo Using base prefix 'C:\\Apps\\Python33' New python executable in foo\Scripts\python.exe Installing setuptools, pip...done. >foo\scripts\activate >pip list pip (1.5.dev1) setuptools (1.3) >pip install Pyreadline Downloading/unpacking Pyreadline Running setup.py (path:C:\Work\personal\foo\build\Pyreadline\setup.py) egg_info for package Pyreadline package init file 'pyreadline\configuration\__init__.py' not found (or not a regular file) Not SVN Repository Installing collected packages: Pyreadline Running setup.py install for Pyreadline package init file 'pyreadline\configuration\__init__.py' not found (or not a regular file) Not SVN Repository Successfully installed Pyreadline Cleaning up... >pip list pip (1.5.dev1) pyreadline (2.0) setuptools (1.3) >python Python 3.3.0 (v3.3.0:bd8afb90ebf2, Sep 29 2012, 10:57:17) [MSC v.1600 64 bit (AMD64)] on win32 Type "help", "copyright", "credits" or "license" for more information. >>> import pip Readline internal error Traceback (most recent call last): File "C:\Work\personal\foo\lib\site-packages\pyreadline\console\console.py", line 768, in hook_wrapper_23 res = ensure_str(readline_hook(prompt)) It does not happen with the latest released pip: >python -m pip install -I pip Downloading/unpacking pip Running setup.py (path:C:\Work\personal\foo\build\pip\setup.py) egg_info for package pip Not SVN Repository warning: no files found matching '*.html' under directory 'docs' warning: no previously-included files matching '*.rst' found under directory 'docs\_build' no previously-included directories found matching 'docs\_build\_sources' Installing collected packages: pip Found existing installation: pip 1.5.dev1 Uninstalling pip: Successfully uninstalled pip Running setup.py install for pip Not SVN Repository warning: no files found matching '*.html' under directory 'docs' warning: no previously-included files matching '*.rst' found under directory 'docs\_build' no previously-included directories found matching 'docs\_build\_sources' Installing pip-3.3-script.py script to C:\Work\personal\foo\Scripts Installing pip-3.3.exe script to C:\Work\personal\foo\Scripts Installing pip-script.py script to C:\Work\personal\foo\Scripts Installing pip.exe script to C:\Work\personal\foo\Scripts Successfully installed pip Cleaning up... >pip list pip (1.4.1) pyreadline (2.0) setuptools (1.3) >python Python 3.3.0 (v3.3.0:bd8afb90ebf2, Sep 29 2012, 10:57:17) [MSC v.1600 64 bit (AMD64)] on win32 Type "help", "copyright", "credits" or "license" for more information. >>> import pip >>> My original thought was that it was a bad interaction with the colorama package (included in pip develop) but I did my best to comment out all uses of pip._vendor.colorama and I *think* I succeeded, and I still got the traceback. There's a few too many load-time and import-time hooks involved for me to be absolutely sure, of course (and the Python process falls over, so I can't do any useful postmortem debugging to see what's loaded...) Paul From francois.deheeger at free.fr Fri Nov 22 09:23:56 2013 From: francois.deheeger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Deheeger?=) Date: Fri, 22 Nov 2013 15:23:56 +0100 Subject: [IPython-dev] title and markdown cell alignment opinion... Message-ID: Hi developers, I got into the habit of watching the dev meeting on thursday evening and found it very interesting (nice teasing on widget stuffs !). (Am i addicted to real TV ?) Yesterday, I felt concern about the first points, and wanted to share my view on that.. So here is just another point of view about title and makdown cells alignment... if it helps. For the title stuff, i pretty agree that it needs a kind of H0 to define the document title. To solve this, i choose to start all my notebooks with a markdown cell that contains this HTML code: """ # title Fran?ois Deheeger   |   DTO/SIM   |    2013-10-22 last update 2013-11-22
""" see attached png. I am parsing this first cell differently in nbconvert to build LaTeX documents and get the title displayed as a title. In "presentation mode" or in "edition mode", this cell looks like an H1 title but the bar insists on the fact that it is not a standard H1. And i do not use the H1 type of cell. On markdown cells alignment, i really like the proposal of having an indent. I am using this kind of indent in my custom.css and found it more readable (My custom.css is a mix between BMFH notebooks and some other pieces of examples from this mailing list.) I think that it's a quite common feeling that texts are more readable when not too wide, and so i reduce also the width of those md cells. I do not feel too hard to see separation between code output and md cell, but this is maybe because of the font choice. (well i have to admit that i am not a big fan of the line separation... ) This was just the point of view of an IPython notebook addict (it becomes my OS since a year). Thanks for making real your vision, we are enjoying it. -- Fran?ois Deheeger -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: nb_head.png Type: image/png Size: 21625 bytes Desc: not available URL: From takowl at gmail.com Fri Nov 22 12:29:50 2013 From: takowl at gmail.com (Thomas Kluyver) Date: Fri, 22 Nov 2013 09:29:50 -0800 Subject: [IPython-dev] title and markdown cell alignment opinion... In-Reply-To: References: Message-ID: Hi Francois, Thanks for the points, we'll take them into consideration. Feel free to log into our development chat room ( http://www.hipchat.com/ghtNzvmfC ) during the meetings, and you can chime in as we're discussing things. We all have that room open during the meetings. Best wishes, Thomas On 22 November 2013 06:23, Fran?ois Deheeger wrote: > Hi developers, > > I got into the habit of watching the dev meeting on thursday evening and > found it very interesting (nice teasing on widget stuffs !). (Am i addicted > to real TV ?) > > Yesterday, I felt concern about the first points, and wanted to share my > view on that.. So here is just another point of view about title and > makdown cells alignment... if it helps. > > For the title stuff, i pretty agree that it needs a kind of H0 to define > the document title. To solve this, i choose to start all my notebooks with > a markdown cell that contains this HTML code: > > """ > > # title > > Fran?ois > Deheeger   |   DTO/SIM   |    > 2013-10-22 > > last update 2013-11-22 > >
> > """ > > see attached png. > > I am parsing this first cell differently in nbconvert to build LaTeX > documents and get the title displayed as a title. > In "presentation mode" or in "edition mode", this cell looks like an H1 > title but the bar insists on the fact that it is not a standard H1. And i > do not use the H1 type of cell. > > > On markdown cells alignment, i really like the proposal of having an > indent. I am using this kind of indent in my custom.css and found it more > readable (My custom.css is a mix between BMFH notebooks and some other > pieces of examples from this mailing list.) > I think that it's a quite common feeling that texts are more readable when > not too wide, and so i reduce also the width of those md cells. > > I do not feel too hard to see separation between code output and md cell, > but this is maybe because of the font choice. (well i have to admit that i > am not a big fan of the line separation... ) > > This was just the point of view of an IPython notebook addict (it becomes > my OS since a year). Thanks for making real your vision, we are enjoying it. > > -- > Fran?ois Deheeger > > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ronena at gmail.com Sun Nov 24 09:34:09 2013 From: ronena at gmail.com (Ronen Abravanel) Date: Sun, 24 Nov 2013 16:34:09 +0200 Subject: [IPython-dev] IPython and matplotlib - Working with both inline plots and plots-in-a-window Message-ID: Hello all, I'm using IPython and matplotlib and like the fact that most of my plots are inline. But once in a while, I want to zoom into a plot, pan it a bit and so on. Is there any way (on stable version of IPython \ matplotlib ) to replot a plot to a new window (\ different backend?) Any ideas? Thanks, Ronen. -------------- next part -------------- An HTML attachment was scrubbed... URL: From bussonniermatthias at gmail.com Sun Nov 24 09:53:21 2013 From: bussonniermatthias at gmail.com (Matthias BUSSONNIER) Date: Sun, 24 Nov 2013 15:53:21 +0100 Subject: [IPython-dev] IPython and matplotlib - Working with both inline plots and plots-in-a-window In-Reply-To: References: Message-ID: <5C564664-F2FC-440F-826B-42DD86595366@gmail.com> Le 24 nov. 2013 ? 15:34, Ronen Abravanel a ?crit : > Hello all, > > I'm using IPython and matplotlib and like the fact that most of my plots are inline. But once in a while, I want to zoom into a plot, pan it a bit and so on. > > Is there any way (on stable version of IPython \ matplotlib ) to replot a plot to a new window (\ different backend?) %matlpotlib qt/osx/wx/inline... allow to switch backend on recent enough version of both IPython and MPL. With non inline backend gca()/ gcf() should grap and show the current graph inline. -- M --- %matplotlib [gui] Set up matplotlib to work interactively. This function lets you activate matplotlib interactive support at any point during an IPython session. It does not import anything into the interactive namespace. If you are using the inline matplotlib backend for embedded figures, you can adjust its behavior via the %config magic:: # enable SVG figures, necessary for SVG+XHTML export in the qtconsole In [1]: %config InlineBackend.figure_format = 'svg' # change the behavior of closing all figures at the end of each # execution (cell), or allowing reuse of active figures across # cells: In [2]: %config InlineBackend.close_figures = False Examples -------- In this case, where the MPL default is TkAgg:: In [2]: %matplotlib Using matplotlib backend: TkAgg But you can explicitly request a different backend:: In [3]: %matplotlib qt positional arguments: gui Name of the matplotlib backend to use ('gtk', 'inline', 'osx', 'qt', 'qt4', 'tk', 'wx'). If given, the corresponding matplotlib backend is used, otherwise it will be matplotlib's default (which you can set in your matplotlib config file). From ronena at gmail.com Sun Nov 24 11:24:07 2013 From: ronena at gmail.com (Ronen Abravanel) Date: Sun, 24 Nov 2013 18:24:07 +0200 Subject: [IPython-dev] IPython and matplotlib - Working with both inline plots and plots-in-a-window In-Reply-To: <5C564664-F2FC-440F-826B-42DD86595366@gmail.com> References: <5C564664-F2FC-440F-826B-42DD86595366@gmail.com> Message-ID: Thanks a lot! this will save a lot of effort. On Sun, Nov 24, 2013 at 4:53 PM, Matthias BUSSONNIER < bussonniermatthias at gmail.com> wrote: > > Le 24 nov. 2013 ? 15:34, Ronen Abravanel a ?crit : > > > Hello all, > > > > I'm using IPython and matplotlib and like the fact that most of my plots > are inline. But once in a while, I want to zoom into a plot, pan it a bit > and so on. > > > > Is there any way (on stable version of IPython \ matplotlib ) to replot > a plot to a new window (\ different backend?) > > %matlpotlib qt/osx/wx/inline... > allow to switch backend on recent enough version of both IPython and MPL. > > With non inline backend gca()/ gcf() should grap and show the current > graph inline. > -- > M > > > --- > %matplotlib [gui] > > Set up matplotlib to work interactively. > > This function lets you activate matplotlib interactive support > at any point during an IPython session. > It does not import anything into the interactive namespace. > > If you are using the inline matplotlib backend for embedded figures, > you can adjust its behavior via the %config magic:: > > # enable SVG figures, necessary for SVG+XHTML export in the qtconsole > In [1]: %config InlineBackend.figure_format = 'svg' > > # change the behavior of closing all figures at the end of each > # execution (cell), or allowing reuse of active figures across > # cells: > In [2]: %config InlineBackend.close_figures = False > > Examples > -------- > In this case, where the MPL default is TkAgg:: > > In [2]: %matplotlib > Using matplotlib backend: TkAgg > > But you can explicitly request a different backend:: > > In [3]: %matplotlib qt > > positional arguments: > gui Name of the matplotlib backend to use ('gtk', 'inline', 'osx', 'qt', > 'qt4', 'tk', 'wx'). If given, the corresponding matplotlib backend > is > used, otherwise it will be matplotlib's default (which you can set > in > your matplotlib config file). > > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > -------------- next part -------------- An HTML attachment was scrubbed... URL: From kasper.peeters at phi-sci.com Mon Nov 25 13:50:13 2013 From: kasper.peeters at phi-sci.com (Kasper Peeters) Date: Mon, 25 Nov 2013 18:50:13 +0000 Subject: [IPython-dev] integrating with Cadabra computer algebra system Message-ID: <20131125185013.6f5d3686@tigger> Hi, I am the author of Cadabra, an open source computer algebra system for tensor computations in field theory, http://cadabra.phi-sci.com/ This system currently has its own notebook-based GUI and its own (restricted) programming language. I am considering switching the language to Python, and also thinking of using IPython for the GUI, to take some of the load off my shoulders. However, I would like the notebook interface to retain the capability of accepting pure Cadabra expressions into an input cell, and have those evaluated using Cadabra rather than Python. I know IPython has the ability to evaluate input cells which contain e.g. R code, but that always seems to require a magic '%' bit at the beginning of a cell. So my question: is there a good way to extend the functionality of the IPython notebook interface so that it allows me to set the input cell type to non-python in a graphical way (e.g. pick from a menu), or perhaps depending on some clever scanning of the input? If not, would you be open to patches from my end? Anyone else here having struggled with a similar problem? (e.g. to use the IPython interface to drive a different computer algebra system and not wanting to put '%'s all over the place?). Thanks for the help. Cheers, Kasper From gmbecker at ucdavis.edu Mon Nov 25 14:11:41 2013 From: gmbecker at ucdavis.edu (Gabriel Becker) Date: Mon, 25 Nov 2013 11:11:41 -0800 Subject: [IPython-dev] integrating with Cadabra computer algebra system In-Reply-To: <20131125185013.6f5d3686@tigger> References: <20131125185013.6f5d3686@tigger> Message-ID: I am not a core IPython developer, but if I understand you correctly what you want is a Cadabra kernel. Which kernel is used when running IPython dictates how code is handled. The (AFAIK) up-to-date and authorative specification for the design of the kernel architecture and what a kernel needs to accept and return is here: http://ipython.org/ipython-doc/stable/development/messaging.html#messaging Other than that, thispurports to explain the construction of a new IPython kernel in the context of node.js, but I am not the author and make no claims as to its currentness or correctness: There is also the IJulia kernel, which I believe was constructed by the Julia folks in direct collaboration with (some part of) the IPython team: https://github.com/JuliaLang/IJulia.jl Hope that helps. ~G On Mon, Nov 25, 2013 at 10:50 AM, Kasper Peeters wrote: > Hi, > > I am the author of Cadabra, an open source computer algebra system for > tensor computations in field theory, > > http://cadabra.phi-sci.com/ > > This system currently has its own notebook-based GUI and its own > (restricted) programming language. I am considering switching the > language to Python, and also thinking of using IPython for the GUI, to > take some of the load off my shoulders. > > However, I would like the notebook interface to retain the capability of > accepting pure Cadabra expressions into an input cell, and have those > evaluated using Cadabra rather than Python. I know IPython has the > ability to evaluate input cells which contain e.g. R code, but that > always seems to require a magic '%' bit at the beginning of a cell. > > So my question: is there a good way to extend the functionality of the > IPython notebook interface so that it allows me to set the input cell > type to non-python in a graphical way (e.g. pick from a menu), or > perhaps depending on some clever scanning of the input? If not, would > you be open to patches from my end? Anyone else here having struggled > with a similar problem? (e.g. to use the IPython interface to drive a > different computer algebra system and not wanting to put '%'s all over > the place?). > > Thanks for the help. > > Cheers, > Kasper > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > -- Gabriel Becker Graduate Student Statistics Department University of California, Davis -------------- next part -------------- An HTML attachment was scrubbed... URL: From takowl at gmail.com Mon Nov 25 14:39:54 2013 From: takowl at gmail.com (Thomas Kluyver) Date: Mon, 25 Nov 2013 11:39:54 -0800 Subject: [IPython-dev] integrating with Cadabra computer algebra system In-Reply-To: <20131125185013.6f5d3686@tigger> References: <20131125185013.6f5d3686@tigger> Message-ID: Hi Kasper, On 25 November 2013 10:50, Kasper Peeters wrote: > However, I would like the notebook interface to retain the capability of > accepting pure Cadabra expressions into an input cell, and have those > evaluated using Cadabra rather than Python. I know IPython has the > ability to evaluate input cells which contain e.g. R code, but that > always seems to require a magic '%' bit at the beginning of a cell. > > So my question: is there a good way to extend the functionality of the > IPython notebook interface so that it allows me to set the input cell > type to non-python in a graphical way (e.g. pick from a menu), or > perhaps depending on some clever scanning of the input? If not, would > you be open to patches from my end? Anyone else here having struggled > with a similar problem? (e.g. to use the IPython interface to drive a > different computer algebra system and not wanting to put '%'s all over > the place?). I think that should be possible. If nothing else, it would be possible to write a JS shim which stored a Cadabra flag in cell-level metadata, and prepended '%%cadabra' when it sent those cells for execution. I hope we can work out something a bit neater than that, though. Best wishes, Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Mon Nov 25 14:40:35 2013 From: fperez.net at gmail.com (Fernando Perez) Date: Mon, 25 Nov 2013 11:40:35 -0800 Subject: [IPython-dev] integrating with Cadabra computer algebra system In-Reply-To: References: <20131125185013.6f5d3686@tigger> Message-ID: One more reference... The author of IHaskell, Andrew Gibiansky, did a terrific job explaining the design of the system as he reverse-engineered it from our docs and the code of IJulia. I highly recommend this to anyone wanting to implement a new kernel for the IPython architecture: http://andrew.gibiansky.com/blog/ipython/ipython-kernels/ Cheers, f On Mon, Nov 25, 2013 at 11:11 AM, Gabriel Becker wrote: > I am not a core IPython developer, but if I understand you correctly what > you want is a Cadabra kernel. > > Which kernel is used when running IPython dictates how code is handled. > > The (AFAIK) up-to-date and authorative specification for the design of the > kernel architecture and what a kernel needs to accept and return is here: > http://ipython.org/ipython-doc/stable/development/messaging.html#messaging > > Other than that, thispurports to explain the construction of a new IPython kernel in the context > of node.js, but I am not the author and make no claims as to its > currentness or correctness: > > There is also the IJulia kernel, which I believe was constructed by the > Julia folks in direct collaboration with (some part of) the IPython team: > https://github.com/JuliaLang/IJulia.jl > > Hope that helps. > ~G > > > On Mon, Nov 25, 2013 at 10:50 AM, Kasper Peeters < > kasper.peeters at phi-sci.com> wrote: > >> Hi, >> >> I am the author of Cadabra, an open source computer algebra system for >> tensor computations in field theory, >> >> http://cadabra.phi-sci.com/ >> >> This system currently has its own notebook-based GUI and its own >> (restricted) programming language. I am considering switching the >> language to Python, and also thinking of using IPython for the GUI, to >> take some of the load off my shoulders. >> >> However, I would like the notebook interface to retain the capability of >> accepting pure Cadabra expressions into an input cell, and have those >> evaluated using Cadabra rather than Python. I know IPython has the >> ability to evaluate input cells which contain e.g. R code, but that >> always seems to require a magic '%' bit at the beginning of a cell. >> >> So my question: is there a good way to extend the functionality of the >> IPython notebook interface so that it allows me to set the input cell >> type to non-python in a graphical way (e.g. pick from a menu), or >> perhaps depending on some clever scanning of the input? If not, would >> you be open to patches from my end? Anyone else here having struggled >> with a similar problem? (e.g. to use the IPython interface to drive a >> different computer algebra system and not wanting to put '%'s all over >> the place?). >> >> Thanks for the help. >> >> Cheers, >> Kasper >> _______________________________________________ >> IPython-dev mailing list >> IPython-dev at scipy.org >> http://mail.scipy.org/mailman/listinfo/ipython-dev >> > > > > -- > Gabriel Becker > Graduate Student > Statistics Department > University of California, Davis > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > > -- Fernando Perez (@fperez_org; http://fperez.org) fperez.net-at-gmail: mailing lists only (I ignore this when swamped!) fernando.perez-at-berkeley: contact me here for any direct mail -------------- next part -------------- An HTML attachment was scrubbed... URL: From ra092767 at ime.unicamp.br Tue Nov 26 13:22:24 2013 From: ra092767 at ime.unicamp.br (Raniere Silva) Date: Tue, 26 Nov 2013 16:22:24 -0200 Subject: [IPython-dev] Trying to diff ipynb after converting it to HTML Message-ID: <20131126182224.GA3701@pupunha> Hi all, I'm trying to get a diff of two ipynb files after converting it to HTML. The script I'm using can be found at https://github.com/r-gaia-cs/diffipynb but the div.input_area.box-flex1 are not show properly (no line breaks) in the output. You can found the HTML output from the test instructions of the README in http://www.ime.unicamp.br/~ra092767/diffipynb/diff.html. Any one know how I can fix it? Thanks, Raniere P.S.: If you want to know why I'm need this, go to https://github.com/swcarpentry/bc/issues/87. From bussonniermatthias at gmail.com Tue Nov 26 14:48:58 2013 From: bussonniermatthias at gmail.com (Matthias BUSSONNIER) Date: Tue, 26 Nov 2013 20:48:58 +0100 Subject: [IPython-dev] Trying to diff ipynb after converting it to HTML In-Reply-To: <20131126182224.GA3701@pupunha> References: <20131126182224.GA3701@pupunha> Message-ID: Le 26 nov. 2013 ? 19:22, Raniere Silva a ?crit : > Hi all, > > I'm trying to get a diff of two ipynb files after converting it to HTML. The > script I'm using can be found at https://github.com/r-gaia-cs/diffipynb but the > div.input_area.box-flex1 are not show properly (no line breaks) in the output. > > You can found the HTML output from the test instructions of the README in > http://www.ime.unicamp.br/~ra092767/diffipynb/diff.html. > > Any one know how I can fix it? You should ask on lxml ML, but you are apparently not the only one : https://mailman-mail5.webfaction.com/pipermail/lxml/2013-July/006914.html -- M > > Thanks, > Raniere > > P.S.: If you want to know why I'm need this, go to > https://github.com/swcarpentry/bc/issues/87. > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev From mark.voorhies at ucsf.edu Tue Nov 26 16:11:31 2013 From: mark.voorhies at ucsf.edu (Mark Voorhies) Date: Tue, 26 Nov 2013 13:11:31 -0800 Subject: [IPython-dev] Trying to diff ipynb after converting it to HTML In-Reply-To: References: <20131126182224.GA3701@pupunha> Message-ID: <52950E83.3060602@ucsf.edu> On 11/26/2013 11:48 AM, Matthias BUSSONNIER wrote: > > Le 26 nov. 2013 ? 19:22, Raniere Silva a ?crit : > >> Hi all, >> >> I'm trying to get a diff of two ipynb files after converting it to HTML. The >> script I'm using can be found at https://github.com/r-gaia-cs/diffipynb but the >> div.input_area.box-flex1 are not show properly (no line breaks) in the output. >> >> You can found the HTML output from the test instructions of the README in >> http://www.ime.unicamp.br/~ra092767/diffipynb/diff.html. >> >> Any one know how I can fix it? > > You should ask on lxml ML, > but you are apparently not the only one : > > https://mailman-mail5.webfaction.com/pipermail/lxml/2013-July/006914.html > Looks like all of the relevant code is in the self-contained lxml/html/diff.py The loss of information (newlines, etc.) happens when tokenizing the input for a word-based diff (e.g., like "git diff --word-diff"): ---- end_whitespace_re = re.compile(r'[ \t\n\r]$') ... def split_words(text): """ Splits some text into words. Includes trailing whitespace (one space) on each word when appropriate. """ if not text or not text.strip(): return [] words = [w + ' ' for w in text.strip().split()] if not end_whitespace_re.search(text): words[-1] = words[-1][:-1] return words start_whitespace_re = re.compile(r'^[ \t\n\r]') ---- So it might be reasonable to write a local module that imports most of the methods from diff.py and customizes split_words and the top level method(s) for 
 elements (either just removing
the white space stripping or switching to line-based diff for 
).

htmldiff appears to be written under the assumption of poorly-formed input, so another approach
(given good markup from IPython) would be to try to use lxml to extract more structure from the
inputs (rather than falling back on regexes).

--Mark




From doc at doconnel.f9.co.uk  Wed Nov 27 05:53:06 2013
From: doc at doconnel.f9.co.uk (Derek O'Connell)
Date: Wed, 27 Nov 2013 10:53:06 +0000
Subject: [IPython-dev] widget/interact modules?
Message-ID: <5295CF12.3060507@doconnel.f9.co.uk>

Which repo contains the widget & interact modules?


From damianavila at gmail.com  Wed Nov 27 06:00:20 2013
From: damianavila at gmail.com (=?ISO-8859-1?Q?Dami=E1n_Avila?=)
Date: Wed, 27 Nov 2013 09:00:20 -0200
Subject: [IPython-dev] widget/interact modules?
In-Reply-To: <5295CF12.3060507@doconnel.f9.co.uk>
References: <5295CF12.3060507@doconnel.f9.co.uk>
Message-ID: 

Currently this PR contains the widget stuff:
https://github.com/ipython/ipython/pull/4374


2013/11/27 Derek O'Connell 

> Which repo contains the widget & interact modules?
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>



-- 
Dami?n Avila
Scientific Python Developer
Quantitative Finance Analyst
Statistics, Biostatistics and Econometrics Consultant
Biochemist
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From doc at doconnel.f9.co.uk  Wed Nov 27 06:59:58 2013
From: doc at doconnel.f9.co.uk (Derek O'Connell)
Date: Wed, 27 Nov 2013 11:59:58 +0000
Subject: [IPython-dev] widget/interact modules?
In-Reply-To: 
References: <5295CF12.3060507@doconnel.f9.co.uk>
	
Message-ID: <5295DEBE.9030003@doconnel.f9.co.uk>

Thanks Dami?n. I followed advice here...

https://gist.github.com/piscisaureus/3342247

...and did:

git fetch origin
git checkout pr/4374
git submodule update
pip install -e ".[notebook]"

The server starts without warnings but the dashboard reports "Error 
connecting to server". Also if I fire up a console and try "from 
IPython.html.widgets.interact ..." I also get as far as "widgets", still 
no "interact" module. Any suggestions?

On 27/11/13 11:00, Dami?n Avila wrote:
> Currently this PR contains the widget stuff: 
> https://github.com/ipython/ipython/pull/4374
>
>
> 2013/11/27 Derek O'Connell  >
>
>     Which repo contains the widget & interact modules?
>     _______________________________________________
>     IPython-dev mailing list
>     IPython-dev at scipy.org 
>     http://mail.scipy.org/mailman/listinfo/ipython-dev
>
>
>
>
> -- 
> Dami?n Avila
> Scientific Python Developer
> Quantitative Finance Analyst
> Statistics, Biostatistics and Econometrics Consultant
> Biochemist
>
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev



From damianavila at gmail.com  Wed Nov 27 08:09:45 2013
From: damianavila at gmail.com (=?ISO-8859-1?Q?Dami=E1n_Avila?=)
Date: Wed, 27 Nov 2013 11:09:45 -0200
Subject: [IPython-dev] widget/interact modules?
In-Reply-To: <5295DEBE.9030003@doconnel.f9.co.uk>
References: <5295CF12.3060507@doconnel.f9.co.uk>
	
	<5295DEBE.9030003@doconnel.f9.co.uk>
Message-ID: 

First, try cleaning you browser cache... and reload the dashboard and try
again...

BTW, I usually use piscisaureus approach... just I have installed IPython
master before fetching branches:

python setupegg.py develop

and then I follow:

git fetch origin
git checkout pr/4374
git submodule update

And the examples at /ipython/examples/widgets are working for me...

I am not used to pip install -e ".[notebook]", but I think it does the same
as develop...




2013/11/27 Derek O'Connell 

> Thanks Dami?n. I followed advice here...
>
> https://gist.github.com/piscisaureus/3342247
>
> ...and did:
>
> git fetch origin
> git checkout pr/4374
> git submodule update
> pip install -e ".[notebook]"
>
> The server starts without warnings but the dashboard reports "Error
> connecting to server". Also if I fire up a console and try "from
> IPython.html.widgets.interact ..." I also get as far as "widgets", still
> no "interact" module. Any suggestions?
>
> On 27/11/13 11:00, Dami?n Avila wrote:
> > Currently this PR contains the widget stuff:
> > https://github.com/ipython/ipython/pull/4374
> >
> >
> > 2013/11/27 Derek O'Connell  > >
> >
> >     Which repo contains the widget & interact modules?
> >     _______________________________________________
> >     IPython-dev mailing list
> >     IPython-dev at scipy.org 
> >     http://mail.scipy.org/mailman/listinfo/ipython-dev
> >
> >
> >
> >
> > --
> > Dami?n Avila
> > Scientific Python Developer
> > Quantitative Finance Analyst
> > Statistics, Biostatistics and Econometrics Consultant
> > Biochemist
> >
> >
> > _______________________________________________
> > IPython-dev mailing list
> > IPython-dev at scipy.org
> > http://mail.scipy.org/mailman/listinfo/ipython-dev
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>



-- 
Dami?n Avila
Scientific Python Developer
Quantitative Finance Analyst
Statistics, Biostatistics and Econometrics Consultant
Biochemist
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From ehmatthes at gmail.com  Wed Nov 27 10:53:32 2013
From: ehmatthes at gmail.com (Eric Matthes)
Date: Wed, 27 Nov 2013 06:53:32 -0900
Subject: [IPython-dev] Installing ipython notebook in virtualenv on mac
Message-ID: 

Hello,

I am trying to help a colleague get IPython Notebook running on a mac. I
use Ubuntu, and I've lost track of everything I did to get IPython running
on my machine. Right now, I can get IPython Notebook working in a virtual
environment by running the following commands:

    ~$ mkdir test_ipython
    ~$ cd test_ipython
    ~/test_ipython$ virtualenv -p python3.3 venv
    ~/test_ipython$ source venv/bin/activate
    (venv)~/test_ipython$ pip install ipython[all]==1.1.0

After this, I can run `ipython --version` and I see `1.1.0`, and I can run
`ipython notebook` successfully from an activated virtual environment.

My colleague is having issues, I believe, with pyzmq. Can someone clarify
what is wrong with these instructions, or what he needs to do to his
overall system to make these instructions work? He has python2.7 and
python3.3 installed on his system.

Also, I am quite new to the list, and if this kind of question really
belongs on Stack Overflow please let me know. I've been hunting around for
a while, but there are so many installation instructions for various
versions of IPython, and so many dependencies that I get lost pretty
quickly in the existing documentation.

Thank you,

Eric Matthes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From mcburton at umich.edu  Wed Nov 27 11:09:56 2013
From: mcburton at umich.edu (mcburton)
Date: Wed, 27 Nov 2013 11:09:56 -0500
Subject: [IPython-dev] Installing ipython notebook in virtualenv on mac
In-Reply-To: 
References: 
Message-ID: 

Eric,

While this doesn't quite answer your specific questions about virtualenv on
a mac, you might check out the GitHub repository for the 2nd edition of *Mining
the Social Web. *Instead of walking users through installing Python and
IPython, the author has them install a linux virtual machine and uses
Vagrant to automatically build an IPython environment.

There is a screencast and instructions on how to set-up the virtual
development environment. I like this approach because it sidesteps the
problem of installing IPython and its dependencies on the local machine.
*https://github.com/ptwobrussell/Mining-the-Social-Web-2nd-Edition#the-mining-the-social-web-virtual-machine
*

If both of you use a VM then you can have the exact same development
environment regardless of your host operating system (linux/win/osx).

Just putting this out as a possible alternative to your challenge.

--
mcb

On Wed, Nov 27, 2013 at 10:53 AM, Eric Matthes  wrote:

> Hello,
>
> I am trying to help a colleague get IPython Notebook running on a mac. I
> use Ubuntu, and I've lost track of everything I did to get IPython running
> on my machine. Right now, I can get IPython Notebook working in a virtual
> environment by running the following commands:
>
>     ~$ mkdir test_ipython
>     ~$ cd test_ipython
>     ~/test_ipython$ virtualenv -p python3.3 venv
>     ~/test_ipython$ source venv/bin/activate
>     (venv)~/test_ipython$ pip install ipython[all]==1.1.0
>
> After this, I can run `ipython --version` and I see `1.1.0`, and I can run
> `ipython notebook` successfully from an activated virtual environment.
>
> My colleague is having issues, I believe, with pyzmq. Can someone clarify
> what is wrong with these instructions, or what he needs to do to his
> overall system to make these instructions work? He has python2.7 and
> python3.3 installed on his system.
>
> Also, I am quite new to the list, and if this kind of question really
> belongs on Stack Overflow please let me know. I've been hunting around for
> a while, but there are so many installation instructions for various
> versions of IPython, and so many dependencies that I get lost pretty
> quickly in the existing documentation.
>
> Thank you,
>
> Eric Matthes
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From wirth.jason at gmail.com  Wed Nov 27 11:14:49 2013
From: wirth.jason at gmail.com (Jason Wirth)
Date: Wed, 27 Nov 2013 10:14:49 -0600
Subject: [IPython-dev] Installing ipython notebook in virtualenv on mac
In-Reply-To: 
References: 
	
Message-ID: 

I would skip virtualenv and go with conda environments. It'll play nicer.
Also you can get set up Anaconda.

On Wednesday, November 27, 2013, mcburton wrote:

> Eric,
>
> While this doesn't quite answer your specific questions about virtualenv
> on a mac, you might check out the GitHub repository for the 2nd edition of *Mining
> the Social Web. *Instead of walking users through installing Python and
> IPython, the author has them install a linux virtual machine and uses
> Vagrant to automatically build an IPython environment.
>
> There is a screencast and instructions on how to set-up the virtual
> development environment. I like this approach because it sidesteps the
> problem of installing IPython and its dependencies on the local machine.
> *https://github.com/ptwobrussell/Mining-the-Social-Web-2nd-Edition#the-mining-the-social-web-virtual-machine
> *
>
> If both of you use a VM then you can have the exact same development
> environment regardless of your host operating system (linux/win/osx).
>
> Just putting this out as a possible alternative to your challenge.
>
> --
> mcb
>
> On Wed, Nov 27, 2013 at 10:53 AM, Eric Matthes 
> > wrote:
>
>> Hello,
>>
>> I am trying to help a colleague get IPython Notebook running on a mac. I
>> use Ubuntu, and I've lost track of everything I did to get IPython running
>> on my machine. Right now, I can get IPython Notebook working in a virtual
>> environment by running the following commands:
>>
>>     ~$ mkdir test_ipython
>>     ~$ cd test_ipython
>>     ~/test_ipython$ virtualenv -p python3.3 venv
>>     ~/test_ipython$ source venv/bin/activate
>>     (venv)~/test_ipython$ pip install ipython[all]==1.1.0
>>
>> After this, I can run `ipython --version` and I see `1.1.0`, and I can
>> run `ipython notebook` successfully from an activated virtual environment.
>>
>> My colleague is having issues, I believe, with pyzmq. Can someone clarify
>> what is wrong with these instructions, or what he needs to do to his
>> overall system to make these instructions work? He has python2.7 and
>> python3.3 installed on his system.
>>
>> Also, I am quite new to the list, and if this kind of question really
>> belongs on Stack Overflow please let me know. I've been hunting around for
>> a while, but there are so many installation instructions for various
>> versions of IPython, and so many dependencies that I get lost pretty
>> quickly in the existing documentation.
>>
>> Thank you,
>>
>> Eric Matthes
>>
>> _______________________________________________
>> IPython-dev mailing list
>> IPython-dev at scipy.org > 'IPython-dev at scipy.org');>
>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>>
>>
>

-- 

-- 
Jason Wirth
    213.675.5294
    wirth.jason at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From ehmatthes at gmail.com  Wed Nov 27 11:23:18 2013
From: ehmatthes at gmail.com (Eric Matthes)
Date: Wed, 27 Nov 2013 07:23:18 -0900
Subject: [IPython-dev] Installing ipython notebook in virtualenv on mac
In-Reply-To: 
References: 
	
Message-ID: 

I had thought about this approach, but I thought working in a virtualenv
would be easier than installing an entire vm. I'll check with the person
I'm working with, and ask his preference.

Eric


On Wed, Nov 27, 2013 at 7:09 AM, mcburton  wrote:

> Eric,
>
> While this doesn't quite answer your specific questions about virtualenv
> on a mac, you might check out the GitHub repository for the 2nd edition of *Mining
> the Social Web. *Instead of walking users through installing Python and
> IPython, the author has them install a linux virtual machine and uses
> Vagrant to automatically build an IPython environment.
>
> There is a screencast and instructions on how to set-up the virtual
> development environment. I like this approach because it sidesteps the
> problem of installing IPython and its dependencies on the local machine.
> *https://github.com/ptwobrussell/Mining-the-Social-Web-2nd-Edition#the-mining-the-social-web-virtual-machine
> *
>
> If both of you use a VM then you can have the exact same development
> environment regardless of your host operating system (linux/win/osx).
>
> Just putting this out as a possible alternative to your challenge.
>
> --
> mcb
>
> On Wed, Nov 27, 2013 at 10:53 AM, Eric Matthes wrote:
>
>> Hello,
>>
>> I am trying to help a colleague get IPython Notebook running on a mac. I
>> use Ubuntu, and I've lost track of everything I did to get IPython running
>> on my machine. Right now, I can get IPython Notebook working in a virtual
>> environment by running the following commands:
>>
>>     ~$ mkdir test_ipython
>>     ~$ cd test_ipython
>>     ~/test_ipython$ virtualenv -p python3.3 venv
>>     ~/test_ipython$ source venv/bin/activate
>>     (venv)~/test_ipython$ pip install ipython[all]==1.1.0
>>
>> After this, I can run `ipython --version` and I see `1.1.0`, and I can
>> run `ipython notebook` successfully from an activated virtual environment.
>>
>> My colleague is having issues, I believe, with pyzmq. Can someone clarify
>> what is wrong with these instructions, or what he needs to do to his
>> overall system to make these instructions work? He has python2.7 and
>> python3.3 installed on his system.
>>
>> Also, I am quite new to the list, and if this kind of question really
>> belongs on Stack Overflow please let me know. I've been hunting around for
>> a while, but there are so many installation instructions for various
>> versions of IPython, and so many dependencies that I get lost pretty
>> quickly in the existing documentation.
>>
>> Thank you,
>>
>> Eric Matthes
>>
>> _______________________________________________
>> IPython-dev mailing list
>> IPython-dev at scipy.org
>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>>
>>
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From rgbkrk at gmail.com  Wed Nov 27 11:47:38 2013
From: rgbkrk at gmail.com (Kyle Kelley)
Date: Wed, 27 Nov 2013 10:47:38 -0600
Subject: [IPython-dev] Installing ipython notebook in virtualenv on mac
In-Reply-To: 
References: 
	
	
Message-ID: 

If your colleague uses brew, I have a set of instructions for installing
ipython notebook+more on Mac OS X in this gist:

https://gist.github.com/rgbkrk/6511241

-- Kyle


On Wed, Nov 27, 2013 at 10:23 AM, Eric Matthes  wrote:

> I had thought about this approach, but I thought working in a virtualenv
> would be easier than installing an entire vm. I'll check with the person
> I'm working with, and ask his preference.
>
> Eric
>
>
> On Wed, Nov 27, 2013 at 7:09 AM, mcburton  wrote:
>
>> Eric,
>>
>> While this doesn't quite answer your specific questions about virtualenv
>> on a mac, you might check out the GitHub repository for the 2nd edition of *Mining
>> the Social Web. *Instead of walking users through installing Python and
>> IPython, the author has them install a linux virtual machine and uses
>> Vagrant to automatically build an IPython environment.
>>
>> There is a screencast and instructions on how to set-up the virtual
>> development environment. I like this approach because it sidesteps the
>> problem of installing IPython and its dependencies on the local machine.
>> *https://github.com/ptwobrussell/Mining-the-Social-Web-2nd-Edition#the-mining-the-social-web-virtual-machine
>> *
>>
>> If both of you use a VM then you can have the exact same development
>> environment regardless of your host operating system (linux/win/osx).
>>
>> Just putting this out as a possible alternative to your challenge.
>>
>> --
>> mcb
>>
>> On Wed, Nov 27, 2013 at 10:53 AM, Eric Matthes wrote:
>>
>>> Hello,
>>>
>>> I am trying to help a colleague get IPython Notebook running on a mac. I
>>> use Ubuntu, and I've lost track of everything I did to get IPython running
>>> on my machine. Right now, I can get IPython Notebook working in a virtual
>>> environment by running the following commands:
>>>
>>>     ~$ mkdir test_ipython
>>>     ~$ cd test_ipython
>>>     ~/test_ipython$ virtualenv -p python3.3 venv
>>>     ~/test_ipython$ source venv/bin/activate
>>>     (venv)~/test_ipython$ pip install ipython[all]==1.1.0
>>>
>>> After this, I can run `ipython --version` and I see `1.1.0`, and I can
>>> run `ipython notebook` successfully from an activated virtual environment.
>>>
>>> My colleague is having issues, I believe, with pyzmq. Can someone
>>> clarify what is wrong with these instructions, or what he needs to do to
>>> his overall system to make these instructions work? He has python2.7 and
>>> python3.3 installed on his system.
>>>
>>> Also, I am quite new to the list, and if this kind of question really
>>> belongs on Stack Overflow please let me know. I've been hunting around for
>>> a while, but there are so many installation instructions for various
>>> versions of IPython, and so many dependencies that I get lost pretty
>>> quickly in the existing documentation.
>>>
>>> Thank you,
>>>
>>> Eric Matthes
>>>
>>> _______________________________________________
>>> IPython-dev mailing list
>>> IPython-dev at scipy.org
>>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>>>
>>>
>>
>> _______________________________________________
>> IPython-dev mailing list
>> IPython-dev at scipy.org
>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>>
>>
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From burkhard at ualberta.ca  Wed Nov 27 12:15:34 2013
From: burkhard at ualberta.ca (Burkhard Ritter)
Date: Wed, 27 Nov 2013 18:15:34 +0100
Subject: [IPython-dev] Installing ipython notebook in virtualenv on mac
In-Reply-To: 
References: 
	
	
	
Message-ID: 

MacPorts also has IPython packages and they are working just fine for me.

Cheers,
Burkhard

On Wed, Nov 27, 2013 at 5:47 PM, Kyle Kelley  wrote:
> If your colleague uses brew, I have a set of instructions for installing
> ipython notebook+more on Mac OS X in this gist:
>
> https://gist.github.com/rgbkrk/6511241
>
> -- Kyle
>
>
> On Wed, Nov 27, 2013 at 10:23 AM, Eric Matthes  wrote:
>>
>> I had thought about this approach, but I thought working in a virtualenv
>> would be easier than installing an entire vm. I'll check with the person I'm
>> working with, and ask his preference.
>>
>> Eric
>>
>>
>> On Wed, Nov 27, 2013 at 7:09 AM, mcburton  wrote:
>>>
>>> Eric,
>>>
>>> While this doesn't quite answer your specific questions about virtualenv
>>> on a mac, you might check out the GitHub repository for the 2nd edition of
>>> Mining the Social Web. Instead of walking users through installing Python
>>> and IPython, the author has them install a linux virtual machine and uses
>>> Vagrant to automatically build an IPython environment.
>>>
>>> There is a screencast and instructions on how to set-up the virtual
>>> development environment. I like this approach because it sidesteps the
>>> problem of installing IPython and its dependencies on the local machine.
>>>
>>> https://github.com/ptwobrussell/Mining-the-Social-Web-2nd-Edition#the-mining-the-social-web-virtual-machine
>>>
>>> If both of you use a VM then you can have the exact same development
>>> environment regardless of your host operating system (linux/win/osx).
>>>
>>> Just putting this out as a possible alternative to your challenge.
>>>
>>> --
>>> mcb
>>>
>>> On Wed, Nov 27, 2013 at 10:53 AM, Eric Matthes 
>>> wrote:
>>>>
>>>> Hello,
>>>>
>>>> I am trying to help a colleague get IPython Notebook running on a mac. I
>>>> use Ubuntu, and I've lost track of everything I did to get IPython running
>>>> on my machine. Right now, I can get IPython Notebook working in a virtual
>>>> environment by running the following commands:
>>>>
>>>>     ~$ mkdir test_ipython
>>>>     ~$ cd test_ipython
>>>>     ~/test_ipython$ virtualenv -p python3.3 venv
>>>>     ~/test_ipython$ source venv/bin/activate
>>>>     (venv)~/test_ipython$ pip install ipython[all]==1.1.0
>>>>
>>>> After this, I can run `ipython --version` and I see `1.1.0`, and I can
>>>> run `ipython notebook` successfully from an activated virtual environment.
>>>>
>>>> My colleague is having issues, I believe, with pyzmq. Can someone
>>>> clarify what is wrong with these instructions, or what he needs to do to his
>>>> overall system to make these instructions work? He has python2.7 and
>>>> python3.3 installed on his system.
>>>>
>>>> Also, I am quite new to the list, and if this kind of question really
>>>> belongs on Stack Overflow please let me know. I've been hunting around for a
>>>> while, but there are so many installation instructions for various versions
>>>> of IPython, and so many dependencies that I get lost pretty quickly in the
>>>> existing documentation.
>>>>
>>>> Thank you,
>>>>
>>>> Eric Matthes
>>>>
>>>> _______________________________________________
>>>> IPython-dev mailing list
>>>> IPython-dev at scipy.org
>>>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>>>>
>>>
>>>
>>> _______________________________________________
>>> IPython-dev mailing list
>>> IPython-dev at scipy.org
>>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>>>
>>
>>
>> _______________________________________________
>> IPython-dev mailing list
>> IPython-dev at scipy.org
>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>>
>
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>


From takowl at gmail.com  Wed Nov 27 13:22:33 2013
From: takowl at gmail.com (Thomas Kluyver)
Date: Wed, 27 Nov 2013 10:22:33 -0800
Subject: [IPython-dev] widget/interact modules?
In-Reply-To: <5295DEBE.9030003@doconnel.f9.co.uk>
References: <5295CF12.3060507@doconnel.f9.co.uk>
	
	<5295DEBE.9030003@doconnel.f9.co.uk>
Message-ID: 

On 27 November 2013 03:59, Derek O'Connell  wrote:

> I also get as far as "widgets", still
> no "interact" module. Any suggestions?
>

That's the widgets PR, interact is another separate branch on top of that.
This will get simpler as we merge these.

Thomas
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From benjaminrk at gmail.com  Wed Nov 27 13:37:52 2013
From: benjaminrk at gmail.com (MinRK)
Date: Wed, 27 Nov 2013 10:37:52 -0800
Subject: [IPython-dev] Installing ipython notebook in virtualenv on mac
In-Reply-To: 
References: 
	
	
	
	
Message-ID: 

And pyzmq has binaries, so you can install it without compiling (on Windows
and OS X) with:

easy_install pyzmq

pip install --use-wheel pyzmq

(depending on pip versions, the wheel may only be found on Windows)




On Wed, Nov 27, 2013 at 9:15 AM, Burkhard Ritter wrote:

> MacPorts also has IPython packages and they are working just fine for me.
>
> Cheers,
> Burkhard
>
> On Wed, Nov 27, 2013 at 5:47 PM, Kyle Kelley  wrote:
> > If your colleague uses brew, I have a set of instructions for installing
> > ipython notebook+more on Mac OS X in this gist:
> >
> > https://gist.github.com/rgbkrk/6511241
> >
> > -- Kyle
> >
> >
> > On Wed, Nov 27, 2013 at 10:23 AM, Eric Matthes 
> wrote:
> >>
> >> I had thought about this approach, but I thought working in a virtualenv
> >> would be easier than installing an entire vm. I'll check with the
> person I'm
> >> working with, and ask his preference.
> >>
> >> Eric
> >>
> >>
> >> On Wed, Nov 27, 2013 at 7:09 AM, mcburton  wrote:
> >>>
> >>> Eric,
> >>>
> >>> While this doesn't quite answer your specific questions about
> virtualenv
> >>> on a mac, you might check out the GitHub repository for the 2nd
> edition of
> >>> Mining the Social Web. Instead of walking users through installing
> Python
> >>> and IPython, the author has them install a linux virtual machine and
> uses
> >>> Vagrant to automatically build an IPython environment.
> >>>
> >>> There is a screencast and instructions on how to set-up the virtual
> >>> development environment. I like this approach because it sidesteps the
> >>> problem of installing IPython and its dependencies on the local
> machine.
> >>>
> >>>
> https://github.com/ptwobrussell/Mining-the-Social-Web-2nd-Edition#the-mining-the-social-web-virtual-machine
> >>>
> >>> If both of you use a VM then you can have the exact same development
> >>> environment regardless of your host operating system (linux/win/osx).
> >>>
> >>> Just putting this out as a possible alternative to your challenge.
> >>>
> >>> --
> >>> mcb
> >>>
> >>> On Wed, Nov 27, 2013 at 10:53 AM, Eric Matthes 
> >>> wrote:
> >>>>
> >>>> Hello,
> >>>>
> >>>> I am trying to help a colleague get IPython Notebook running on a
> mac. I
> >>>> use Ubuntu, and I've lost track of everything I did to get IPython
> running
> >>>> on my machine. Right now, I can get IPython Notebook working in a
> virtual
> >>>> environment by running the following commands:
> >>>>
> >>>>     ~$ mkdir test_ipython
> >>>>     ~$ cd test_ipython
> >>>>     ~/test_ipython$ virtualenv -p python3.3 venv
> >>>>     ~/test_ipython$ source venv/bin/activate
> >>>>     (venv)~/test_ipython$ pip install ipython[all]==1.1.0
> >>>>
> >>>> After this, I can run `ipython --version` and I see `1.1.0`, and I can
> >>>> run `ipython notebook` successfully from an activated virtual
> environment.
> >>>>
> >>>> My colleague is having issues, I believe, with pyzmq. Can someone
> >>>> clarify what is wrong with these instructions, or what he needs to do
> to his
> >>>> overall system to make these instructions work? He has python2.7 and
> >>>> python3.3 installed on his system.
> >>>>
> >>>> Also, I am quite new to the list, and if this kind of question really
> >>>> belongs on Stack Overflow please let me know. I've been hunting
> around for a
> >>>> while, but there are so many installation instructions for various
> versions
> >>>> of IPython, and so many dependencies that I get lost pretty quickly
> in the
> >>>> existing documentation.
> >>>>
> >>>> Thank you,
> >>>>
> >>>> Eric Matthes
> >>>>
> >>>> _______________________________________________
> >>>> IPython-dev mailing list
> >>>> IPython-dev at scipy.org
> >>>> http://mail.scipy.org/mailman/listinfo/ipython-dev
> >>>>
> >>>
> >>>
> >>> _______________________________________________
> >>> IPython-dev mailing list
> >>> IPython-dev at scipy.org
> >>> http://mail.scipy.org/mailman/listinfo/ipython-dev
> >>>
> >>
> >>
> >> _______________________________________________
> >> IPython-dev mailing list
> >> IPython-dev at scipy.org
> >> http://mail.scipy.org/mailman/listinfo/ipython-dev
> >>
> >
> >
> > _______________________________________________
> > IPython-dev mailing list
> > IPython-dev at scipy.org
> > http://mail.scipy.org/mailman/listinfo/ipython-dev
> >
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From doc at doconnel.f9.co.uk  Wed Nov 27 19:25:41 2013
From: doc at doconnel.f9.co.uk (Derek O'Connell)
Date: Thu, 28 Nov 2013 00:25:41 +0000
Subject: [IPython-dev] widget/interact modules?
In-Reply-To: 
References: <5295CF12.3060507@doconnel.f9.co.uk>
	
	<5295DEBE.9030003@doconnel.f9.co.uk>
	
Message-ID: <52968D85.4060906@doconnel.f9.co.uk>

Thanks, got it.

On 27/11/13 18:22, Thomas Kluyver wrote:
> On 27 November 2013 03:59, Derek O'Connell  > wrote:
>
>     I also get as far as "widgets", still
>     no "interact" module. Any suggestions?
>
>
> That's the widgets PR, interact is another separate branch on top of 
> that. This will get simpler as we merge these.
>
> Thomas
>
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev



From zvoros at gmail.com  Fri Nov 29 10:05:14 2013
From: zvoros at gmail.com (=?ISO-8859-1?Q?Zolt=E1n_V=F6r=F6s?=)
Date: Fri, 29 Nov 2013 16:05:14 +0100
Subject: [IPython-dev] defining latex commands in notebook
Message-ID: <5298AD2A.5080305@gmail.com>

Hi all,

I was trying to define some latex commands in the notebook, and there it 
works fine. I run into trouble, however, when I try to convert it to pdf 
via LaTeX. I found that in such a case one could place the new commands 
in a raw cell 
(http://python.6.x6.nabble.com/newcommand-in-Markdown-cells-and-nbconver-td5037469.html), 
and the statement was that those cells would be passed unchanged into 
the latex source file. But in my case, that's not what happens. During 
conversion, lines in the raw cell are prepended by a hash symbol, and 
that obviously throws LaTeX off.

So, my question is, whether this has changed since Oct 30 (the date of 
the above-referred posting), or I am just missing something trivial.

Thanks,

Zolt?n


From bussonniermatthias at gmail.com  Fri Nov 29 10:14:55 2013
From: bussonniermatthias at gmail.com (Matthias BUSSONNIER)
Date: Fri, 29 Nov 2013 16:14:55 +0100
Subject: [IPython-dev] defining latex commands in notebook
In-Reply-To: <5298AD2A.5080305@gmail.com>
References: <5298AD2A.5080305@gmail.com>
Message-ID: <3F65B9E7-48DE-4A8C-BAC2-16432F6CE88E@gmail.com>


Le 29 nov. 2013 ? 16:05, Zolt?n V?r?s a ?crit :

> Hi all,
> 
> I was trying to define some latex commands in the notebook, and there it 
> works fine. I run into trouble, however, when I try to convert it to pdf 
> via LaTeX. I found that in such a case one could place the new commands 
> in a raw cell 
> (http://python.6.x6.nabble.com/newcommand-in-Markdown-cells-and-nbconver-td5037469.html), 
> and the statement was that those cells would be passed unchanged into 
> the latex source file. But in my case, that's not what happens. During 
> conversion, lines in the raw cell are prepended by a hash symbol, and 
> that obviously throws LaTeX off.
> 
> So, my question is, whether this has changed since Oct 30 (the date of 
> the above-referred posting), or I am just missing something trivial.

It should not have changed (yet)
https://github.com/ipython/ipython/pull/4306
--
M

-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From damontallen at gmail.com  Fri Nov 29 10:16:56 2013
From: damontallen at gmail.com (Damon Allen)
Date: Fri, 29 Nov 2013 10:16:56 -0500
Subject: [IPython-dev] defining latex commands in notebook
In-Reply-To: <3F65B9E7-48DE-4A8C-BAC2-16432F6CE88E@gmail.com>
References: <5298AD2A.5080305@gmail.com>
	<3F65B9E7-48DE-4A8C-BAC2-16432F6CE88E@gmail.com>
Message-ID: 

I haven't used functions like this but I have found if you have a LaTex
command defined in a raw cell at the beginning of a notebook it does note
get hashed out.  I just put the following two lines in a raw cell and
converted it to check:

$\newcommand{\Mt}[1]{\mathbf{#1}}$
> \pagestyle{plain}
>

followed by a simple code cell, 2+2, and it survived the conversion.

Damon



On Fri, Nov 29, 2013 at 10:14 AM, Matthias BUSSONNIER <
bussonniermatthias at gmail.com> wrote:

>
> Le 29 nov. 2013 ? 16:05, Zolt?n V?r?s a ?crit :
>
> Hi all,
>
> I was trying to define some latex commands in the notebook, and there it
> works fine. I run into trouble, however, when I try to convert it to pdf
> via LaTeX. I found that in such a case one could place the new commands
> in a raw cell
> (
> http://python.6.x6.nabble.com/newcommand-in-Markdown-cells-and-nbconver-td5037469.html),
>
> and the statement was that those cells would be passed unchanged into
> the latex source file. But in my case, that's not what happens. During
> conversion, lines in the raw cell are prepended by a hash symbol, and
> that obviously throws LaTeX off.
>
> So, my question is, whether this has changed since Oct 30 (the date of
> the above-referred posting), or I am just missing something trivial.
>
>
> It should not have changed (yet)
> https://github.com/ipython/ipython/pull/4306
> --
> M
>
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From zvoros at gmail.com  Fri Nov 29 12:51:59 2013
From: zvoros at gmail.com (=?ISO-8859-1?Q?Zolt=E1n_V=F6r=F6s?=)
Date: Fri, 29 Nov 2013 18:51:59 +0100
Subject: [IPython-dev] defining latex commands in notebook
In-Reply-To: 
References: <5298AD2A.5080305@gmail.com>	<3F65B9E7-48DE-4A8C-BAC2-16432F6CE88E@gmail.com>
	
Message-ID: <5298D43F.5010706@gmail.com>


On 29/11/13 16:16, Damon Allen wrote:
> I haven't used functions like this but I have found if you have a 
> LaTex command defined in a raw cell at the beginning of a notebook it 
> does note get hashed out.  I just put the following two lines in a raw 
> cell and converted it to check:
>
>     $\newcommand{\Mt}[1]{\mathbf{#1}}$
>     \pagestyle{plain}
>
> followed by a simple code cell, 2+2, and it survived the conversion.
>
> Damon
This doesn't work in my case. The lines in the raw cell are still hashed:

# $\newcommand{\Mt}[1]{\mathbf{#1}}$
# \pagestyle{plain}


Is there something in the templates that produces this? I am using the 
latest ipython from master.

Cheers,
Zolt?n
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From damontallen at gmail.com  Fri Nov 29 13:58:02 2013
From: damontallen at gmail.com (Damon Allen)
Date: Fri, 29 Nov 2013 13:58:02 -0500
Subject: [IPython-dev] defining latex commands in notebook
In-Reply-To: <5298D43F.5010706@gmail.com>
References: <5298AD2A.5080305@gmail.com>
	<3F65B9E7-48DE-4A8C-BAC2-16432F6CE88E@gmail.com>
	
	<5298D43F.5010706@gmail.com>
Message-ID: 

My version and yours may be different.  I'm just using the 1.1
releaserather than the
current 2.0 version.   Alternatively, it might be an issue
with sphinx.  I have installed sphinx and tornado using pip rather than
using what is in the repositories.

if you are using Python3 you might have to install get-pip.py from git to
have pip-3.2.

curl -O https://raw.github.com/pypa/pip/master/contrib/get-pip.py
>

Damon

On Fri, Nov 29, 2013 at 12:51 PM, Zolt?n V?r?s  wrote:

>
> On 29/11/13 16:16, Damon Allen wrote:
>
>  I haven't used functions like this but I have found if you have a LaTex
> command defined in a raw cell at the beginning of a notebook it does note
> get hashed out.  I just put the following two lines in a raw cell and
> converted it to check:
>
> $\newcommand{\Mt}[1]{\mathbf{#1}}$
>> \pagestyle{plain}
>>
>
> followed by a simple code cell, 2+2, and it survived the conversion.
>
>  Damon
>
> This doesn't work in my case. The lines in the raw cell are still hashed:
>
> # $\newcommand{\Mt}[1]{\mathbf{#1}}$
> # \pagestyle{plain}
>
>
> Is there something in the templates that produces this? I am using the
> latest ipython from master.
>
> Cheers,
> Zolt?n
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From zvoros at gmail.com  Fri Nov 29 14:14:09 2013
From: zvoros at gmail.com (=?ISO-8859-1?Q?Zolt=E1n_V=F6r=F6s?=)
Date: Fri, 29 Nov 2013 20:14:09 +0100
Subject: [IPython-dev] defining latex commands in notebook
In-Reply-To: 
References: <5298AD2A.5080305@gmail.com>	<3F65B9E7-48DE-4A8C-BAC2-16432F6CE88E@gmail.com>		<5298D43F.5010706@gmail.com>
	
Message-ID: <5298E781.3030900@gmail.com>


On 29/11/13 19:58, Damon Allen wrote:
> My version and yours may be different.  I'm just using the 1.1 release 
>  rather than the current 
> 2.0 version.   Alternatively, it might be an issue with sphinx.  I 
> have installed sphinx and tornado using pip rather than using what is 
> in the repositories.
I suspect sphinx then. (I guess, tornado doesn't matter in this case.) 
My version is 1.1.3. Is yours the same?

Zolt?n

-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From damontallen at gmail.com  Fri Nov 29 15:33:41 2013
From: damontallen at gmail.com (Damon Allen)
Date: Fri, 29 Nov 2013 15:33:41 -0500
Subject: [IPython-dev] defining latex commands in notebook
In-Reply-To: <5298E781.3030900@gmail.com>
References: <5298AD2A.5080305@gmail.com>
	<3F65B9E7-48DE-4A8C-BAC2-16432F6CE88E@gmail.com>
	
	<5298D43F.5010706@gmail.com>
	
	<5298E781.3030900@gmail.com>
Message-ID: 

I have 1.2b1.

Damon

On Fri, Nov 29, 2013 at 2:14 PM, Zolt?n V?r?s  wrote:

>
> On 29/11/13 19:58, Damon Allen wrote:
>
>  My version and yours may be different.  I'm just using the 1.1 releaserather than the current 2.0 version.   Alternatively, it might be an issue
> with sphinx.  I have installed sphinx and tornado using pip rather than
> using what is in the repositories.
>
> I suspect sphinx then. (I guess, tornado doesn't matter in this case.) My
> version is 1.1.3. Is yours the same?
>
> Zolt?n
>
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From takowl at gmail.com  Fri Nov 29 18:30:52 2013
From: takowl at gmail.com (Thomas Kluyver)
Date: Fri, 29 Nov 2013 15:30:52 -0800
Subject: [IPython-dev] Fwd: [SciTech] ActivePapers Python edition
In-Reply-To: 
References: <21096.55140.14608.59254@Ordinateur-de-Catherine-Konrad.local>
	
Message-ID: 

Have any of us come across ActivePapers? It combines code, data and
documentation into an HDF5 file to be uploaded as supplementary info for a
journal article. It looks like it focuses more on the storage and
reusability than the presentation and interactivity of IPython notebooks -
so there could be interesting opportunities for integration, such as an
IPython notebook backed by code and data stored in this form.

https://bitbucket.org/khinsen/active_papers_py/wiki/Home

Thomas


---------- Forwarded message ----------
From: Konrad Hinsen 
Date: 24 October 2013 01:16
Subject: [SciTech] ActivePapers Python edition
To: scitech at hackingscience.org


Hi everyone,

The Python port of my ActivePapers project has advanced to the point
that I make a public announcement. For an overview, see

   http://bitbucket.org/khinsen/active_papers_py/wiki/Home

For more details and a practical example, see the tutorial:

    http://bitbucket.org/khinsen/active_papers_py/wiki/Tutorial

For two recent publications that come with an ActivePaper as
supplementary material, see

    http://dx.doi.org/10.1063/1.4821598
    http://dx.doi.org/10.1063/1.4823996


In brief, ActivePapers is a framework for doing and publishing
reproducible computational science that addresses three key points
that weren't covered by existing tools:

 - potentially large binary datasets
 - computations spread over multiple machines (desktop, cluster, computing
centre, ...)
 - installation-free use and re-use of published code

The first two points were a requirement for much of my own work, so
the main motivation for ActivePapers was to allow me to not only
preach, but also practice reproducible research.

An ActivePaper is a single file containing any combination of data and
code. It can re-use previously published data and code by references
to other ActivePapers through a DOI. References are downloaded
automatically (for now this works only with Figshare).

The original ActivePapers project
(https://bitbucket.org/khinsen/active_papers) was a proof of concept
for implementing as many desirable features as possible in a single
framework for reproducible research. It used JVM bytecode for code
storage, for reasons of longevity, universality, and secure
execution. For this reason, it was of little practical interest to
most of today's computational science practitioners because most
existing scientific software is not based on the JVM. The Python port
had to remove some of the features of the original project, but is
usable right now with many widely used software packages.

Konrad.
--
---------------------------------------------------------------------
Konrad Hinsen
Centre de Biophysique Mol?culaire, CNRS Orl?ans
Synchrotron Soleil - Division Exp?riences
Saint Aubin - BP 48
91192 Gif sur Yvette Cedex, France
Tel. +33-1 69 35 97 15
E-Mail: research AT khinsen DOT fastmail DOT net
http://dirac.cnrs-orleans.fr/~hinsen/
ORCID: http://orcid.org/0000-0003-0330-9428
Twitter: @khinsen
---------------------------------------------------------------------
_______________________________________________
SciTech mailing list
SciTech at lists.hackingscience.org
http://lists.hackingscience.org/listinfo.cgi/scitech-hackingscience.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From ehmatthes at gmail.com  Sat Nov 30 19:45:45 2013
From: ehmatthes at gmail.com (Eric Matthes)
Date: Sat, 30 Nov 2013 15:45:45 -0900
Subject: [IPython-dev] Running IPython Notebook against Python 2.7 or Python
	3.3
Message-ID: 

Hello,

I have loved working with IPython Notebook, but I've become totally
confused about how I set up my main system. I love how it is set up now; I
can start a virtualenv with ipython, that either uses python 2.7 or python
3.3. But if I try to build a similar setup on a clean install of 12.04 from
virtualbox, I can't replicate my environment.

I am not tied to using virtualenv for ipython, but I do need a reliable
setup that lets me choose to use ipython notebook with either Python 2.7 or
3.3. I've played with virtualenvs, with pip and easy_install, with symlinks
from python and python3.3 directories, with dist-packages and
site-packages, and it's left me pretty confused.

Can someone suggest a straightforward way to set up an Ubuntu 12.04 system
to use IPython Notebook, so that I can choose to have the notebook using a
Python 2.7 or Python 3.3 interpreter at different times? For context, I am
working on some tutorials that reference both 2.7 code samples and 3.3 code
samples, and I'd like to be able to run both. I have an open question on
stackoverflow about installing ipython notebook to a virtualenv, but I
don't need that particular question answered necessarily at this point if
there is a better alternative.

(so question) -
http://stackoverflow.com/questions/20290357/ipython-notebook-in-a-virtualenv-using-python-3-3

Eric Matthes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: