\n \n \n \n \n .... Could we simply prepend a

Thanks for making this available! -Doug > _______________________________________________ > 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 doug.blank at gmail.com Tue Jul 29 09:06:35 2014 From: doug.blank at gmail.com (Doug Blank) Date: Tue, 29 Jul 2014 09:06:35 -0400 Subject: [IPython-dev] How to make a notebook extension In-Reply-To: References: Message-ID: Follow-ups (suggestions, options, overall betterness) available in the associated issue tracker: https://github.com/ipython/ipython/issues/6229 Hopefully will lead to good documentation. -Doug On Sun, Jul 27, 2014 at 1:24 PM, Doug Blank wrote: > Ok, I think I have this figured out, and if the following is right, and > best practice, I'd be glad to put in the docs. > > First, write your JavaScript programs. There should be a main JS file. The > JavaScript file should generally take the form: > > ``` > define( function () { > var load_ipython_extension = function () { > ... > }; > > return { > load_ipython_extension : load_ipython_extension, > }; > }); > ``` > > This function will return a JS object with "load_ipython_extension" > defined to be a function that is called when loading the extension. > > Name this JS file with a unique name that won't collide with other > extensions. For example, use "jones-feature-name.js" where "jones" is the > name of your group, and "feature-name" is the name of what the feature you > are adding/enhancing. > > If there is only the single .js file, then you can easily make a download > for it. > > If more than one file is needed for your extension (additional libraries, > css, etc) it is advised to make a zip file for easy install. The zip file > should have your main JS file in the root. > > To install the zip or js file, use: > > ipython install-nbextension URL > > The files are saved to $(ipython locale)/nbextensions/ > > At this point, you can use your extension by interactively loading it in a > notebook (regardless of kernel, as long as your kernel supports running > javascript); in the native kernel, that is: > > ``` > %%javascript > IPython.load_extensions("jones-feature-name") > ``` > > Your feature is now active, and will be loaded automatically each time you > open this notebook if you leave this cell in the notebook. [Is that right? > How does it work?] > > What if you have an associated CSS file? It is not loaded automatically. > > Add this to your defined JS object: > > ``` > var load_css = function () { > var link = document.createElement("link"); > link.type = "text/css"; > link.rel = "stylesheet"; > link.href = require.toUrl("/nbextensions/jones-feature-name.css"); > console.log(link); > document.getElementsByTagName("head")[0].appendChild(link); > }; > ``` > > and call it in load_ipython_extension. > > Finally, if you want to load your extension automatically (or need to have > your script loaded earlier to handle opening functions) then you can add > this single line to your $(ipython > locale)/profile_XXX/static/custom/custom.js: > > IPython.load_extensions("jones-feature-name"); > > There are more options (having things execute on notebook open, adding > toolbar buttons, menu items), and those are explain here ... > > How does that sound? Did I miss anything? > > -Doug > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From hughesadam87 at gmail.com Tue Jul 29 13:22:16 2014 From: hughesadam87 at gmail.com (Adam Hughes) Date: Tue, 29 Jul 2014 13:22:16 -0400 Subject: [IPython-dev] HTML pandas issue (followup of twitter question) Message-ID: Thanks for answering my twitter question about HTML representations of dataframes. https://twitter.com/hughesadam87 I have a composition class that stores a DataFrame, and uses __getattr__ to pass any attribute call down to the dataframe. Therefore, my object has a __repr__(), _repr_html_() etc... The difference is, if I do "print obj.df", I get the nice html output by default. However, if I do "print obj", I don't get this; I get the standard plain string output. This is weird because I am not overwriting anything. I can sucessfully get my object to print the dataframe html representation by doing: from IPython.core.display import HTML display(HTML(obj.to_html())) What I don't understand is how does the notebook know that a pandas dataframe should display as html by default? How are they communicating? When you print a dataframe, does the notebook look specifically for the _repr_html Since my object has all of the methods of a dataframe (although not publicized), I imagine the notebook might be looking for certain attributes or methods or something, to set this default HTML printing? Hope that makes some sense, sorry. -------------- next part -------------- An HTML attachment was scrubbed... URL: From takowl at gmail.com Tue Jul 29 13:26:43 2014 From: takowl at gmail.com (Thomas Kluyver) Date: Tue, 29 Jul 2014 10:26:43 -0700 Subject: [IPython-dev] HTML pandas issue (followup of twitter question) In-Reply-To: References: Message-ID: On 29 July 2014 10:22, Adam Hughes wrote: > The difference is, if I do "print obj.df", I get the nice html output by > default. That's strange - print should only ever give you the plain string representation. If you just put 'obj.df' or 'obj' on a line by itself at the end of the cell, *without* print, it should display HTML reprs. If you want to display more than one thing per cell, then IPython.display.display(obj) is the rich-repr equivalent of print. Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From hughesadam87 at gmail.com Tue Jul 29 14:28:53 2014 From: hughesadam87 at gmail.com (Adam Hughes) Date: Tue, 29 Jul 2014 14:28:53 -0400 Subject: [IPython-dev] HTML pandas issue (followup of twitter question) In-Reply-To: References:

Message-ID: I'm sorry you're right thomas. I didn't mean print; I meant if I leave it as the last statement in a cell. IE ... df vs. ... myobject In this case, it still just prints plain text for myobject. The only print method that myobject specifically overloads is: def __repr__(self): return self._df.__repr__() Since otherwise it would default to python objects' print statements. Could this be the problem? I can try to make a self-contained reproducible example if you think this is a bug of sorts. On Tue, Jul 29, 2014 at 1:26 PM, Thomas Kluyver wrote: > On 29 July 2014 10:22, Adam Hughes wrote: > >> The difference is, if I do "print obj.df", I get the nice html output by >> default. > > > That's strange - print should only ever give you the plain string > representation. If you just put 'obj.df' or 'obj' on a line by itself at > the end of the cell, *without* print, it should display HTML reprs. > > If you want to display more than one thing per cell, then > IPython.display.display(obj) is the rich-repr equivalent of print. > > 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 Jul 29 14:32:14 2014 From: takowl at gmail.com (Thomas Kluyver) Date: Tue, 29 Jul 2014 11:32:14 -0700 Subject: [IPython-dev] HTML pandas issue (followup of twitter question) In-Reply-To: References:

Message-ID: On 29 July 2014 11:28, Adam Hughes wrote: > The only print method that myobject specifically overloads is: > > def __repr__(self): > Didn't you say you also define _repr_html_()? It's the presence of _repr_html_() on your object that lets IPython give it the HTML display. Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From tritemio at gmail.com Tue Jul 29 14:36:49 2014 From: tritemio at gmail.com (Antonino Ingargiola) Date: Tue, 29 Jul 2014 11:36:49 -0700 Subject: [IPython-dev] IPython notebook widgets for interactive matplotlib figures Message-ID: HI to all, I'm following up to this SO question: http://stackoverflow.com/questions/24961674/ipython-notebook-widgets-for-matplotlib-interactivity To summarize the problem is that I want to "modify" a plot via some notebook widgets, for example adding a cursor (however can can think about any plot tweak including adding/removing data). Putting the following code in a single cell: fig, ax = plt.subplots() ax.plot([3,1,2,4,0,5,3,2,0,2,4]) vline = ax.axvline(1) hline = ax.axhline(0.5) def set_cursor(x, y): vline.set_xdata((x, x)) hline.set_ydata((y, y)) display(fig) interact(set_cursor, x=(1, 9, 0.01), y=(0, 5, 0.01)) results in 2 figures being displayed: one by the standard notebook machinery in the Out field and one by the interact call in the widgets space (between the code cell and the output). Moving the slider makes the first plot to disappear. Is this the expected behavior? Am I doing something wrong? Is there a way I can inhibit the first figure display? Thanks, Antonio -------------- next part -------------- An HTML attachment was scrubbed... URL: From hughesadam87 at gmail.com Tue Jul 29 15:03:15 2014 From: hughesadam87 at gmail.com (Adam Hughes) Date: Tue, 29 Jul 2014 15:03:15 -0400 Subject: [IPython-dev] HTML pandas issue (followup of twitter question) In-Reply-To: References:

Message-ID: AHHHHHHHHH I FORGOT THE TRAILING UNDERSCORE!!!!!!!! Thanks Thomas. My life is complete now. On Tue, Jul 29, 2014 at 2:32 PM, Thomas Kluyver wrote: > On 29 July 2014 11:28, Adam Hughes wrote: > >> The only print method that myobject specifically overloads is: >> >> def __repr__(self): >> > > Didn't you say you also define _repr_html_()? It's the presence of > _repr_html_() on your object that lets IPython give it the HTML display. > > 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 fperez.net at gmail.com Tue Jul 29 15:21:07 2014 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 29 Jul 2014 12:21:07 -0700 Subject: [IPython-dev] IPython notebook widgets for interactive matplotlib figures In-Reply-To: References: Message-ID: On Tue, Jul 29, 2014 at 11:36 AM, Antonino Ingargiola wrote: > > Is this the expected behavior? Am I doing something wrong? Is there a way > I can inhibit the first figure display? > It's an unfortunate side effect of IPython's internal logic that tries to auto-display plots so that simple calls like plot(x) work in an intuitive fashion. Since you're manually calling display(), both code paths kick in, and you end up with a duplicate figure. The simple answer is: in every instance where you're going to do manual figure displaying yourself, you can prevent the automatic display by putting plt.close(fig) at the end of your cell, if you have a handle on a specific figure, or more simply (if less elegant) plt.close('all') that will close every open figure regardless how many there are or whether you have a handle on them. We were never able to come up with a solution that would work well in all the naive/simple cases and yet also not have funny side effects in more advanced scenarios like this. In the end, we settled on making the simple cases work as best as possible, and having to make a manual close() call in other cases. It's possible that with a bit more thought a better solution can be found, this was done early on (before the notebook, all that logic was designed initially for the Qt console). We had so many things to do that we moved on with this current solution, but it may be worth revisiting to see if we can come up with better heuristics or ideas. 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 tritemio at gmail.com Tue Jul 29 16:06:17 2014 From: tritemio at gmail.com (Antonino Ingargiola) Date: Tue, 29 Jul 2014 13:06:17 -0700 Subject: [IPython-dev] IPython notebook widgets for interactive matplotlib figures In-Reply-To: References: Message-ID: On Tue, Jul 29, 2014 at 12:21 PM, Fernando Perez wrote: > > On Tue, Jul 29, 2014 at 11:36 AM, Antonino Ingargiola > wrote: > >> >> Is this the expected behavior? Am I doing something wrong? Is there a way >> I can inhibit the first figure display? >> > > It's an unfortunate side effect of IPython's internal logic that tries to > auto-display plots so that simple calls like > > plot(x) > > work in an intuitive fashion. Since you're manually calling display(), > both code paths kick in, and you end up with a duplicate figure. > > The simple answer is: in every instance where you're going to do manual > figure displaying yourself, you can prevent the automatic display by putting > > plt.close(fig) > > at the end of your cell, if you have a handle on a specific figure > Ouch! So simple! Thanks for the solution. I didn't tried close() because I though that closing the figure would destroy it. It turns out that it is not and the following interact call works as expected. > We were never able to come up with a solution that would work well in all > the naive/simple cases and yet also not have funny side effects in more > advanced scenarios like this. In the end, we settled on making the simple > cases work as best as possible, and having to make a manual close() call in > other cases. It's possible that with a bit more thought a better solution > can be found, this was done early on (before the notebook, all that logic > was designed initially for the Qt console). We had so many things to do > that we moved on with this current solution, but it may be worth revisiting > to see if we can come up with better heuristics or ideas. > Just my 2c: why don't providing an API function like "show_figure(False)". At this moment it will just call close(). In the future if you change implementation the API function will stay the same. Best, Antonio -------------- next part -------------- An HTML attachment was scrubbed... URL: From msg2mw at gmail.com Tue Jul 29 16:24:20 2014 From: msg2mw at gmail.com (Mike Witt) Date: Tue, 29 Jul 2014 13:24:20 -0700 Subject: [IPython-dev] Problem using JSAnimation Message-ID: <1406665460.6801.58@Backie> I've never used JSAnimation before. I just tried installing it, and every time I try to do anything with it, it wants to import things from matplotlib.animation that don't seem to be there (at least on my system). I wonder, is there some piece of matplolib that I'm missing? I should mention that I've never used matplotlib.animation either. In the past I've created the frames by hand and put them together with convert, ffmpeg, or something like that. In [1]: from JSAnimation import IPython_display --------------------------------------------------------------------------- ImportError Traceback (most recent call last) in () ----> 1 from JSAnimation import IPython_display /usr/local/lib/python2.7/dist-packages/JSAnimation/__init__.py in () ----> 1 from .html_writer import HTMLWriter /usr/local/lib/python2.7/dist-packages/JSAnimation/html_writer.py in () 8 else: 9 from io import BytesIO as InMemory ---> 10 from matplotlib.animation import writers, FileMovieWriter 11 from base64 import b64encode 12 ImportError: cannot import name writers From fperez.net at gmail.com Tue Jul 29 16:41:36 2014 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 29 Jul 2014 13:41:36 -0700 Subject: [IPython-dev] IPython notebook widgets for interactive matplotlib figures In-Reply-To: References: Message-ID: On Tue, Jul 29, 2014 at 1:06 PM, Antonino Ingargiola wrote: > Just my 2c: why don't providing an API function like "show_figure(False)". > At this moment it will just call close(). In the future if you change > implementation the API function will stay the same. Not a bad idea... I'd like to mull a little bit on the cleanest way to do this, but that's not a bad solution. If nothing else, you can add that to your local utils for now, and see if it works well enough. def showfig(f): plt.close(f) display(f) is easy enough to add at the top :) 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 Tue Jul 29 16:42:58 2014 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 29 Jul 2014 13:42:58 -0700 Subject: [IPython-dev] Problem using JSAnimation In-Reply-To: <1406665460.6801.58@Backie> References: <1406665460.6801.58@Backie> Message-ID: Check your matplotlib version, it may be too old. I have 1.2.1 and those APIs are there, otherwise you may need to update mpl. On Tue, Jul 29, 2014 at 1:24 PM, Mike Witt wrote: > I've never used JSAnimation before. I just tried installing it, > and every time I try to do anything with it, it wants to import > things from matplotlib.animation that don't seem to be there (at > least on my system). I wonder, is there some piece of matplolib > that I'm missing? > > I should mention that I've never used matplotlib.animation either. > In the past I've created the frames by hand and put them together > with convert, ffmpeg, or something like that. > > > In [1]: from JSAnimation import IPython_display > --------------------------------------------------------------------------- > ImportError Traceback (most recent call > last) > in () > ----> 1 from JSAnimation import IPython_display > > /usr/local/lib/python2.7/dist-packages/JSAnimation/__init__.py in > () > ----> 1 from .html_writer import HTMLWriter > > /usr/local/lib/python2.7/dist-packages/JSAnimation/html_writer.py in > () > 8 else: > 9 from io import BytesIO as InMemory > ---> 10 from matplotlib.animation import writers, FileMovieWriter > 11 from base64 import b64encode > 12 > > ImportError: cannot import name writers > _______________________________________________ > 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 msg2mw at gmail.com Tue Jul 29 16:54:09 2014 From: msg2mw at gmail.com (Mike Witt) Date: Tue, 29 Jul 2014 13:54:09 -0700 Subject: [IPython-dev] Problem using JSAnimation In-Reply-To: (from fperez.net@gmail.com on Tue Jul 29 13:42:58 2014) References: <1406665460.6801.58@Backie> Message-ID: <1406667249.6801.60@Backie> On 07/29/2014 01:42:58 PM, Fernando Perez wrote: > Check your matplotlib version, it may be too old. I have 1.2.1 and > those > APIs are there, otherwise you may need to update mpl. Mine is older. I'm updating. Thanks! -Mike From markbak at gmail.com Tue Jul 29 16:56:03 2014 From: markbak at gmail.com (Mark Bakker) Date: Tue, 29 Jul 2014 22:56:03 +0200 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing Message-ID: We have developed a series of 15 Notebooks for scientists and engineers who want to use Python programming for exploratory computing, scipting, data analysis, and visualization.The Notebooks were developed for an undergraduate class (sophomores) in Civil Engineering. No prior knowledge of computer programming is assumed. Each Notebook covers a specific topic and includes a number of exercises. The exercises should take less than 4 hours to complete for each Notebook. The Notebooks may be viewed at http://mbakker7.github.io/exploratory_computing_with_python/ A link to the GitHub repository is also shown on this page. Hope you find them useful, Mark Bakker TU Delft, The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: From tritemio at gmail.com Tue Jul 29 17:11:06 2014 From: tritemio at gmail.com (Antonino Ingargiola) Date: Tue, 29 Jul 2014 14:11:06 -0700 Subject: [IPython-dev] IPython notebook widgets for interactive matplotlib figures In-Reply-To: References:

Message-ID: On Tue, Jul 29, 2014 at 1:41 PM, Fernando Perez wrote: > > On Tue, Jul 29, 2014 at 1:06 PM, Antonino Ingargiola > wrote: > >> Just my 2c: why don't providing an API function like >> "show_figure(False)". At this moment it will just call close(). In the >> future if you change implementation the API function will stay the same. > > > Not a bad idea... I'd like to mull a little bit on the cleanest way to do > this, but that's not a bad solution. > > If nothing else, you can add that to your local utils for now, and see if > it works well enough. > > def showfig(f): > plt.close(f) > display(f) > > is easy enough to add at the top :) > Yep, I'm doing also a step further. Since the problem was adding interactivity to inline matplotlib plots, I wrapped the necessary bits in a function like this: def add_cursor(fig, ax): plt.close() vline = ax.axvline(1, color='k') hline = ax.axhline(1, color='k') def set_cursor(x, y): vline.set_xdata((x, x)) hline.set_ydata((y, y)) display(fig) interact(set_cursor, x=ax.get_xlim(), y=ax.get_ylim()) and then simply add a call to add_cursor whenever is needed. In this case display() is called by interact(). I will probably write a couple of similar functions for various "interactivity needs". Thanks again, Antonio -------------- next part -------------- An HTML attachment was scrubbed... URL: From satra at mit.edu Tue Jul 29 17:21:07 2014 From: satra at mit.edu (Satrajit Ghosh) Date: Tue, 29 Jul 2014 17:21:07 -0400 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing In-Reply-To: References: Message-ID: thanks mark! this is great and came at an opportune time. i was just looking to create similar stats modules. some related questions to the larger community. - how could we disseminate or find such modules that can be used for teaching, without having to go through every single notebook? - how do we describe the prior knowledge required through a dependency structure? cheers, satra On Tue, Jul 29, 2014 at 4:56 PM, Mark Bakker wrote: > We have developed a series of 15 Notebooks for scientists and engineers > who want to use Python programming for exploratory computing, scipting, > data analysis, and visualization.The Notebooks were developed for an > undergraduate class (sophomores) in Civil Engineering. No prior knowledge > of computer programming is assumed. Each Notebook covers a specific topic > and includes a number of exercises. The exercises should take less than 4 > hours to complete for each Notebook. > > The Notebooks may be viewed at > http://mbakker7.github.io/exploratory_computing_with_python/ > A link to the GitHub repository is also shown on this page. > > Hope you find them useful, > > Mark Bakker > TU Delft, The Netherlands > > _______________________________________________ > 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 fperez.net at gmail.com Tue Jul 29 17:34:11 2014 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 29 Jul 2014 14:34:11 -0700 Subject: [IPython-dev] IPython notebook widgets for interactive matplotlib figures In-Reply-To: References:

Message-ID: On Tue, Jul 29, 2014 at 2:11 PM, Antonino Ingargiola wrote: > def add_cursor(fig, ax): > plt.close() > Make sure you call close on the fig object, so you don't rely on the implicit state management that pyplot does... -- 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 Jul 29 17:35:41 2014 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 29 Jul 2014 14:35:41 -0700 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing In-Reply-To: References:

Message-ID: On Tue, Jul 29, 2014 at 2:21 PM, Satrajit Ghosh wrote: > - how could we disseminate or find such modules that can be used for > teaching, without having to go through every single notebook? > For now, the gallery is the best we have. We plan on putting resources into making nbviewer a much richer platform, but Matthias keep having this silly notion of "finishing a PhD" that constantly gets in the way of real work ;) > - how do we describe the prior knowledge required through a dependency > structure? > Thant sounds like something a semantic web person would say... ;) -- 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 tritemio at gmail.com Tue Jul 29 17:39:23 2014 From: tritemio at gmail.com (Antonino Ingargiola) Date: Tue, 29 Jul 2014 14:39:23 -0700 Subject: [IPython-dev] IPython notebook widgets for interactive matplotlib figures In-Reply-To: References:

Message-ID: On Tue, Jul 29, 2014 at 2:34 PM, Fernando Perez wrote: > > On Tue, Jul 29, 2014 at 2:11 PM, Antonino Ingargiola > wrote: > >> def add_cursor(fig, ax): >> plt.close() >> > > Make sure you call close on the fig object, so you don't rely on the > implicit state management that pyplot does... > Right, thanks! Antonio -------------- next part -------------- An HTML attachment was scrubbed... URL: From bussonniermatthias at gmail.com Wed Jul 30 04:26:37 2014 From: bussonniermatthias at gmail.com (Matthias Bussonnier) Date: Wed, 30 Jul 2014 10:26:37 +0200 Subject: [IPython-dev] [IPython-User] IPython Notebooks and Active Learning (Matthew Brett) In-Reply-To: References:

Message-ID: If I may suggest too, please use header cell instead of hash in markdown. Header cell are used for section when converting to latex and do automatically generate anchors on nbviewer/html so that you can link to subpart: Example : http://nbviewer.ipython.org/url/jdj.mit.edu/~stevenj/IJulia%20Preview.ipynb#Multimedia-display-in-IJulia On Tue, Jul 29, 2014 at 11:16 PM, Matthew Brett wrote: > Hi, > > On Tue, Jul 29, 2014 at 4:50 PM, Mark Bakker wrote: > > We have developed a series of 15 Notebooks for scientists and engineers > who > > want to use Python programming for exploratory computing, scipting, data > > analysis, and visualization. No prior knowledge of computer programming > is > > assumed. Each Notebook covers a specific topic and includes a number of > > exercises. The exercises should take less than 4 hours to complete for > each > > Notebook. > > > > We have developed these Notebooks for an undergraduate class > (sophomores) in > > Civil Engineering. A short survey of the students taking the class (~270 > of > > them) showed that the students really liked the class and learned a lot. > > > > The Notebooks may be viewed at > > http://mbakker7.github.io/exploratory_computing_with_python/ > > A link to the GitHub repository is also shown on this page. > > Thanks a lot for this. From our feedback, our students liked the > notebooks too - e.g. > > "I appreciate the downloadable iPython notebooks with explanations of > what the code is and does - will be a great reference" > > I think there's really no question that the notebooks make running > code examples easier and clearer for the teacher and the student. And > they are indeed a great reference. The question always is - what do > we want to teach? In some cases it's probably enough that the > students get the idea, and running / writing code in the notebooks > helps them get the idea. But the students also implicitly learn that > this is the standard way of working with code, and I personally don't > think we should be teaching that. So, for me at least, I am trying to > find a way to strike a balance between the ease of writing materials / > ease of getting students running code, and the need for teaching solid > working practice. For example, for the next iteration of our course, > I'm thinking of doing a flipped classroom format, with the tutorials > mostly in IPython notebooks, but doing the exercises in class using > text editor and terminal and git. I'd also like to try and teach the > IPython notebook as a great tool for sharing and explaining a > workflow, or developing a tutorial. > > Cheers, > > Matthew > _______________________________________________ > IPython-User mailing list > IPython-User at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-user > -------------- next part -------------- An HTML attachment was scrubbed... URL: From doug.blank at gmail.com Wed Jul 30 07:33:03 2014 From: doug.blank at gmail.com (Doug Blank) Date: Wed, 30 Jul 2014 07:33:03 -0400 Subject: [IPython-dev] [IPython-User] IPython Notebooks and Active Learning (Matthew Brett) In-Reply-To: References:

Message-ID: Excellent project! What is the license for these? It would be great if they were Creative Commons Share with Attribution, or similar: https://creativecommons.org/licenses/by-sa/4.0/ I can imagine using variations of these for a Physics curriculum. On Wed, Jul 30, 2014 at 4:26 AM, Matthias Bussonnier < bussonniermatthias at gmail.com> wrote: > If I may suggest too, please use header cell instead of hash in markdown. > Header cell are used for section when converting to latex and do > automatically generate anchors on nbviewer/html so that you can link to > subpart: > > Example : > > http://nbviewer.ipython.org/url/jdj.mit.edu/~stevenj/IJulia%20Preview.ipynb#Multimedia-display-in-IJulia > Being in the middle of writing similar notebooks, we found that using header cells can be quite painful when it comes to moving sections around. For example, if you want to move a section, one has to individually select, press up/down, for all cells in a section. So, we wrote "Move Section Up"/"Move Section Down" buttons and have included those in calico-document-tools. The tool will move all cells in a section/subsection up or down. You can install it: ipython install-nbextension https://bitbucket.org/ipre/calico/downloads/ calico-document-tools-1.0.zip You can then begin to use them in your notebooks by loading them dynamically: ``` %%javascript IPython.load_extensions("calico-document-tools"); ``` Or, if you want to load them always, put that line in your: $(ipython locate)/profile_default/static/custom/custom.js That collection of tools also contains: section numbering, table of contents, and a beta version of bibliographic support. -Doug > > > On Tue, Jul 29, 2014 at 11:16 PM, Matthew Brett > wrote: > >> Hi, >> >> On Tue, Jul 29, 2014 at 4:50 PM, Mark Bakker wrote: >> > We have developed a series of 15 Notebooks for scientists and engineers >> who >> > want to use Python programming for exploratory computing, scipting, data >> > analysis, and visualization. No prior knowledge of computer programming >> is >> > assumed. Each Notebook covers a specific topic and includes a number of >> > exercises. The exercises should take less than 4 hours to complete for >> each >> > Notebook. >> > >> > We have developed these Notebooks for an undergraduate class >> (sophomores) in >> > Civil Engineering. A short survey of the students taking the class >> (~270 of >> > them) showed that the students really liked the class and learned a lot. >> > >> > The Notebooks may be viewed at >> > http://mbakker7.github.io/exploratory_computing_with_python/ >> > A link to the GitHub repository is also shown on this page. >> >> Thanks a lot for this. From our feedback, our students liked the >> notebooks too - e.g. >> >> "I appreciate the downloadable iPython notebooks with explanations of >> what the code is and does - will be a great reference" >> >> I think there's really no question that the notebooks make running >> code examples easier and clearer for the teacher and the student. And >> they are indeed a great reference. The question always is - what do >> we want to teach? In some cases it's probably enough that the >> students get the idea, and running / writing code in the notebooks >> helps them get the idea. But the students also implicitly learn that >> this is the standard way of working with code, and I personally don't >> think we should be teaching that. So, for me at least, I am trying to >> find a way to strike a balance between the ease of writing materials / >> ease of getting students running code, and the need for teaching solid >> working practice. For example, for the next iteration of our course, >> I'm thinking of doing a flipped classroom format, with the tutorials >> mostly in IPython notebooks, but doing the exercises in class using >> text editor and terminal and git. I'd also like to try and teach the >> IPython notebook as a great tool for sharing and explaining a >> workflow, or developing a tutorial. >> >> Cheers, >> >> Matthew >> _______________________________________________ >> IPython-User mailing list >> IPython-User at scipy.org >> http://mail.scipy.org/mailman/listinfo/ipython-user >> > > > _______________________________________________ > 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 markbak at gmail.com Wed Jul 30 16:57:45 2014 From: markbak at gmail.com (Mark Bakker) Date: Wed, 30 Jul 2014 22:57:45 +0200 Subject: [IPython-dev] [IPython-User] IPython Notebooks and Active Learning (Matthew Brett) Message-ID: Thanks, Doug. Hope you find it useful. Just added the license to the website. And thanks for your pointers about moving header cells around. Will have to try out your tools. Definitely interested in the bibliographic support. Is there any other support around to handle bibliography stuff (I am thinking bibtex-like capabilities...). Keep me informed how your development goes, Mark From: Doug Blank > > Excellent project! What is the license for these? It would be great if they > were Creative Commons Share with Attribution, or similar: > > https://creativecommons.org/licenses/by-sa/4.0/ > > I can imagine using variations of these for a Physics curriculum. > > On Wed, Jul 30, 2014 at 4:26 AM, Matthias Bussonnier < > bussonniermatthias at gmail.com> wrote: > > > If I may suggest too, please use header cell instead of hash in markdown. > > Header cell are used for section when converting to latex and do > > automatically generate anchors on nbviewer/html so that you can link to > > subpart: > > > > Example : > > > > > http://nbviewer.ipython.org/url/jdj.mit.edu/~stevenj/IJulia%20Preview.ipynb#Multimedia-display-in-IJulia > > > > Being in the middle of writing similar notebooks, we found that using > header cells can be quite painful when it comes to moving sections around. > For example, if you want to move a section, one has to individually select, > press up/down, for all cells in a section. > > So, we wrote "Move Section Up"/"Move Section Down" buttons and have > included those in calico-document-tools. The tool will move all cells in a > section/subsection up or down. You can install it: > > ipython install-nbextension https://bitbucket.org/ipre/calico/downloads/ > calico-document-tools-1.0.zip > > You can then begin to use them in your notebooks by loading them > dynamically: > > ``` > %%javascript > IPython.load_extensions("calico-document-tools"); > ``` > > Or, if you want to load them always, put that line in your: > > $(ipython locate)/profile_default/static/custom/custom.js > > That collection of tools also contains: section numbering, table of > contents, and a beta version of bibliographic support. > > -Doug > > > > > > > > On Tue, Jul 29, 2014 at 11:16 PM, Matthew Brett > > > wrote: > > > >> Hi, > >> > >> On Tue, Jul 29, 2014 at 4:50 PM, Mark Bakker wrote: > >> > We have developed a series of 15 Notebooks for scientists and > engineers > >> who > >> > want to use Python programming for exploratory computing, scipting, > data > >> > analysis, and visualization. No prior knowledge of computer > programming > >> is > >> > assumed. Each Notebook covers a specific topic and includes a number > of > >> > exercises. The exercises should take less than 4 hours to complete for > >> each > >> > Notebook. > >> > > >> > We have developed these Notebooks for an undergraduate class > >> (sophomores) in > >> > Civil Engineering. A short survey of the students taking the class > >> (~270 of > >> > them) showed that the students really liked the class and learned a > lot. > >> > > >> > The Notebooks may be viewed at > >> > http://mbakker7.github.io/exploratory_computing_with_python/ > >> > A link to the GitHub repository is also shown on this page. > >> > >> Thanks a lot for this. From our feedback, our students liked the > >> notebooks too - e.g. > >> > >> "I appreciate the downloadable iPython notebooks with explanations of > >> what the code is and does - will be a great reference" > >> > >> I think there's really no question that the notebooks make running > >> code examples easier and clearer for the teacher and the student. And > >> they are indeed a great reference. The question always is - what do > >> we want to teach? In some cases it's probably enough that the > >> students get the idea, and running / writing code in the notebooks > >> helps them get the idea. But the students also implicitly learn that > >> this is the standard way of working with code, and I personally don't > >> think we should be teaching that. So, for me at least, I am trying to > >> find a way to strike a balance between the ease of writing materials / > >> ease of getting students running code, and the need for teaching solid > >> working practice. For example, for the next iteration of our course, > >> I'm thinking of doing a flipped classroom format, with the tutorials > >> mostly in IPython notebooks, but doing the exercises in class using > >> text editor and terminal and git. I'd also like to try and teach the > >> IPython notebook as a great tool for sharing and explaining a > >> workflow, or developing a tutorial. > >> > >> Cheers, > >> > >> Matthew > >> _______________________________________________ > >> IPython-User mailing list > >> IPython-User at scipy.org > >> http://mail.scipy.org/mailman/listinfo/ipython-user > >> > > > > > > _______________________________________________ > > 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: > http://mail.scipy.org/pipermail/ipython-dev/attachments/20140730/0fe0738d/attachment-0001.html > > ------------------------------ > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > > > End of IPython-dev Digest, Vol 126, Issue 66 > ******************************************** > -------------- next part -------------- An HTML attachment was scrubbed... URL: From kikocorreoso at gmail.com Wed Jul 30 18:40:44 2014 From: kikocorreoso at gmail.com (Kiko) Date: Thu, 31 Jul 2014 00:40:44 +0200 Subject: [IPython-dev] Command line tool to publish a notebook on a wordpress blog In-Reply-To: References:

Message-ID: 2014-07-29 11:00 GMT+02:00 Matthias Bussonnier : > Hi > Le 29 juil. 2014 ? 10:07, Kiko a ?crit : > > Hi all, > > I made a small script to publish a notebook directly on a wordpress blog. > The repo can be found here: > https://github.com/Pybonacci/ipy2wp > > It uses nbconvert to convert the notebook to html (maybe I should explore > a conversion to markdown instead of html). I use this to publish on > http://pybonacci.org and it works great with simple notebooks (I have to > check notebooks with more complex structure). It creates a draft on your > wordpress blog so you can check the post before publishing. > > Hi, Matthias. How can I change the basic template with a custom template? Is it possible? Should I use jinja2? markdown? Any thought about that will be very welcome. > > That's really nice, > > It is super agreeable to see people using it as a library and not using > subprocess ! > We've spent quite some time working on that so it make me glad to see > people using it ! > > You should add it to the Wiki on project using IPython/Jupyter ! > > -- > Matthias > > > If someone uses the script please provide some feedback to improve the > tool. If someone experience any problem, please, open an issue in the repo. > If someone has something similar and want to share it would be great to see > how others are solving that. > > If your wordpress blog is older than v 3.5 please check this link: > http://codex.wordpress.org/XML-RPC_Support > > Kind regards. > _______________________________________________ > 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 kikocorreoso at gmail.com Wed Jul 30 18:44:54 2014 From: kikocorreoso at gmail.com (Kiko) Date: Thu, 31 Jul 2014 00:44:54 +0200 Subject: [IPython-dev] Command line tool to publish a notebook on a wordpress blog In-Reply-To: References:

Message-ID: 2014-07-29 14:07 GMT+02:00 Doug Blank : > On Tue, Jul 29, 2014 at 4:07 AM, Kiko wrote: > >> Hi all, >> >> I made a small script to publish a notebook directly on a wordpress blog. >> The repo can be found here: >> https://github.com/Pybonacci/ipy2wp >> >> It uses nbconvert to convert the notebook to html (maybe I should explore >> a conversion to markdown instead of html). I use this to publish on >> http://pybonacci.org and it works great with simple notebooks (I have to >> check notebooks with more complex structure). It creates a draft on your >> wordpress blog so you can check the post before publishing. >> >> If someone uses the script please provide some feedback to improve the >> tool. If someone experience any problem, please, open an issue in the repo. >> If someone has something similar and want to share it would be great to see >> how others are solving that. >> >> If your wordpress blog is older than v 3.5 please check this link: >> http://codex.wordpress.org/XML-RPC_Support >> >> Kind regards. >> > > Very useful! > > Do you have any suggestions regarding how to get IPython's /static etc. > installed so that everything needed (eg, javascript) will load? > > Also, looks like some HTML is being injected into

> > Well, wordpress change some html. One thing would be to use a plugin like 'raw html' or similar. But the basic idea of the tool would be to help in the publication of a notebook in a wordpress blog but some stuff are dependent of your wordpress install (javascript, iframes, code highlight,...). If someone have any ideas about how to improve it I'm all ears (eyes in this case) > Thanks for making this available! > > -Doug > > >> _______________________________________________ >> 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 kikocorreoso at gmail.com Wed Jul 30 18:48:40 2014 From: kikocorreoso at gmail.com (Kiko) Date: Thu, 31 Jul 2014 00:48:40 +0200 Subject: [IPython-dev] Command line tool to publish a notebook on a wordpress blog In-Reply-To: References:

Message-ID: 2014-07-31 0:44 GMT+02:00 Kiko : > > > > 2014-07-29 14:07 GMT+02:00 Doug Blank : > > On Tue, Jul 29, 2014 at 4:07 AM, Kiko wrote: >> >>> Hi all, >>> >>> I made a small script to publish a notebook directly on a wordpress >>> blog. The repo can be found here: >>> https://github.com/Pybonacci/ipy2wp >>> >>> It uses nbconvert to convert the notebook to html (maybe I should >>> explore a conversion to markdown instead of html). I use this to publish on >>> http://pybonacci.org and it works great with simple notebooks (I have >>> to check notebooks with more complex structure). It creates a draft on your >>> wordpress blog so you can check the post before publishing. >>> >>> If someone uses the script please provide some feedback to improve the >>> tool. If someone experience any problem, please, open an issue in the repo. >>> If someone has something similar and want to share it would be great to see >>> how others are solving that. >>> >>> If your wordpress blog is older than v 3.5 please check this link: >>> http://codex.wordpress.org/XML-RPC_Support >>> >>> Kind regards. >>> >> >> By the way, I changed some stuff and the inline images now are uploaded to your wordpress (wp-content/uploads) and then linked in the html uploaded. -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Wed Jul 30 19:30:04 2014 From: fperez.net at gmail.com (Fernando Perez) Date: Wed, 30 Jul 2014 16:30:04 -0700 Subject: [IPython-dev] [IPython-User] IPython Notebooks and Active Learning (Matthew Brett) In-Reply-To: References:

Message-ID: On Wed, Jul 30, 2014 at 4:33 AM, Doug Blank wrote: > Being in the middle of writing similar notebooks, we found that using > header cells can be quite painful when it comes to moving sections around. > For example, if you want to move a section, one has to individually select, > press up/down, for all cells in a section. That is really a big pain point in managing any kind of complex/long notebook. Paul Ivanov recently protoyped out a solution for this: https://github.com/ivanov/nb-cccp I've been using it and it works fairly well, even though it's not really production-ready yet. We in the core team are really trying to get 3.0 out the door, so I'm not sure how much bandwidth Paul will have for this in the coming weeks. But it would be a great tool to finish up, because it's *super* useful. If anyone can pitch in to help Paul out with this code, it would be a great contribution that would make everyone's life a lot easier... 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 lstagner at uci.edu Wed Jul 30 20:29:41 2014 From: lstagner at uci.edu (Luke Stagner) Date: Wed, 30 Jul 2014 17:29:41 -0700 Subject: [IPython-dev] IDL/GDL kernel Message-ID: <53D98DF5.7000100@uci.edu> Hello All, I am a graduate student studying plasma physics. I have made a mostly functional IDL/GDL(just GDL for now) kernel using the new wrapper kernel feature. I borrowed heavily from the bash_kernel example. If it is not too much trouble I need some help with two things. 1. IPython doesn't kill GDL process when shutting down or restarting the kernel. I have defined the do_shutdown to kill the process but that doesn't seem to do anything. 2. Inline Plots: My current method is to have GDL make pngs of the plots and then publishing them. Unfortunately this is not working and I am not sure why. Plasma physics is very much stuck using IDL because of all the legacy software. My hope is that this kernel will act as a gateway drug towards more open programming languages. When IPython 3/Jupyter comes out I am going to try to convince the computer staff at the DIII-D National Ignition Facility to install the multi-user notebook server with many different kernels installed. Thanks for your time -Luke Stagner -------------- next part -------------- An HTML attachment was scrubbed... URL: From mmckerns at caltech.edu Wed Jul 30 20:40:31 2014 From: mmckerns at caltech.edu (Michael McKerns) Date: Wed, 30 Jul 2014 20:40:31 -0400 (EDT) Subject: [IPython-dev] IDL/GDL kernel In-Reply-To: <53D98DF5.7000100@uci.edu> References: <53D98DF5.7000100@uci.edu> Message-ID: <59944.67.186.183.87.1406767231.squirrel@webmail.caltech.edu> Luke, I know it's not what you are asking? but just to throw this into the mix. you might be able to bridge some gaps with one of my old packages, pyIDL. I'm just putting it out there in case you haven't come across it. It's unmaintained, but was fairly feature complete a few years back. I just stopped using IDL. https://pypi.python.org/pypi/pyIDL/ http://www.cacr.caltech.edu/~mmckerns/pyIDL.html http://sourceforge.net/projects/pyidl/ I'm guessing that you could see what I did to stop the IDL processes in my IDL bindings, and maybe then leverage that. > Hello All, > > I am a graduate student studying plasma physics. I have made a mostly > functional IDL/GDL(just GDL for now) kernel > using the new wrapper kernel > feature. I borrowed heavily from the bash_kernel > example. If it is not too > much trouble I need some help with two things. > > 1. IPython doesn't kill GDL process when shutting down or restarting the > kernel. I have defined the do_shutdown > > to kill the process but that doesn't seem to do anything. > 2. Inline Plots: My current method > > is to have GDL make pngs of the plots and then publishing them. > Unfortunately this is not working and I am not sure why. > > Plasma physics is very much stuck using IDL because of all the legacy > software. My hope is that this kernel will act as a gateway drug towards > more open programming languages. When IPython 3/Jupyter comes out I am > going to try to convince the computer staff at the DIII-D National > Ignition Facility to install the multi-user notebook server with many > different kernels installed. > > Thanks for your time > > -Luke Stagner > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > --- Mike McKerns California Institute of Technology http://www.its.caltech.edu/~mmckerns From doug.blank at gmail.com Wed Jul 30 20:42:11 2014 From: doug.blank at gmail.com (Doug Blank) Date: Wed, 30 Jul 2014 20:42:11 -0400 Subject: [IPython-dev] [IPython-User] IPython Notebooks and Active Learning (Matthew Brett) In-Reply-To: References:

Message-ID: On Wed, Jul 30, 2014 at 7:30 PM, Fernando Perez wrote: > > On Wed, Jul 30, 2014 at 4:33 AM, Doug Blank wrote: > >> Being in the middle of writing similar notebooks, we found that using >> header cells can be quite painful when it comes to moving sections around. >> For example, if you want to move a section, one has to individually select, >> press up/down, for all cells in a section. > > > That is really a big pain point in managing any kind of complex/long > notebook. Paul Ivanov recently protoyped out a solution for this: > > https://github.com/ivanov/nb-cccp > > I've been using it and it works fairly well, even though it's not really > production-ready yet. We in the core team are really trying to get 3.0 out > the door, so I'm not sure how much bandwidth Paul will have for this in the > coming weeks. But it would be a great tool to finish up, because it's > *super* useful. > > If anyone can pitch in to help Paul out with this code, it would be a > great contribution that would make everyone's life a lot easier... > Yes, many problems can be solved by being able to select multiple cells, and operate on them. One should be able to make many of the operations that can work on a selected cell to work on a set of selected cells. This extension is really only a temporary workaround for that kind of functionality. And, once there is a method to select multiple cells, it would be really handy to be able to drag and drop those cells where you want them. But, I am really looking forward to a server-based Jupyter setup, with selectable kernels, so I don't want to distract from the big picture :) With the set of extension tools we have now, we are confident that this will work well for a writing course this fall. -Doug > > 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 takowl at gmail.com Wed Jul 30 20:54:42 2014 From: takowl at gmail.com (Thomas Kluyver) Date: Wed, 30 Jul 2014 17:54:42 -0700 Subject: [IPython-dev] IDL/GDL kernel In-Reply-To: <53D98DF5.7000100@uci.edu> References: <53D98DF5.7000100@uci.edu> Message-ID: Hi Luke, On 30 July 2014 17:29, Luke Stagner wrote: > I am a graduate student studying plasma physics. I have made a mostly > functional IDL/GDL(just GDL for now) kernel > using the new wrapper kernel > feature. I borrowed heavily from the bash_kernel > example. If it is not too much > trouble I need some help with two things. > Cool, I'm glad to hear that the API and the example are useful. > 1. IPython doesn't kill GDL process when shutting down or restarting the > kernel. I have defined the do_shutdown > > to kill the process but that doesn't seem to do anything. > That's strange. I've just tried with my bash kernel, which doesn't implement do_shutdown at all, and it does stop the process correctly (the server stops the wrapper process, and that automatically causes the child process to die). I don't know exactly what could be interfering with that, but I'll keep thinking. > 2. Inline Plots: My current method > is > to have GDL make pngs of the plots and then publishing them. Unfortunately > this is not working and I am not sure why. > It should work. Looking at your code, you should switch back to the commented out line that opens and reads the image files, rather than displaying the filenames. If you comment out the rmtree(plot_dir) line, are the images being written as you expect? Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Wed Jul 30 21:20:03 2014 From: fperez.net at gmail.com (Fernando Perez) Date: Wed, 30 Jul 2014 18:20:03 -0700 Subject: [IPython-dev] [IPython-User] IPython Notebooks and Active Learning (Matthew Brett) In-Reply-To: References:

Message-ID: On Wed, Jul 30, 2014 at 5:42 PM, Doug Blank wrote: > Yes, many problems can be solved by being able to select multiple cells, > and operate on them. One should be able to make many of the operations that > can work on a selected cell to work on a set of selected cells. This > extension is really only a temporary workaround for that kind of > functionality. > > And, once there is a method to select multiple cells, it would be really > handy to be able to drag and drop those cells where you want them. > Yup, ultimately that's the UI work that needs to be done. > > But, I am really looking forward to a server-based Jupyter setup, with > selectable kernels, so I don't want to distract from the big picture :) > With the set of extension tools we have now, we are confident that this > will work well for a writing course this fall. > Thanks for your patience, and kudos on the great work you've been doing on your side! 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 lstagner at uci.edu Wed Jul 30 21:32:13 2014 From: lstagner at uci.edu (Luke Stagner) Date: Wed, 30 Jul 2014 18:32:13 -0700 Subject: [IPython-dev] IDL/GDL kernel In-Reply-To: References: Message-ID: <53D99C9D.4070900@uci.edu> Thomas, I am glad I am not the only one stumped by this. I have done what you have suggested and I verified that it is producing the .png files. If I were to hazard a guess I would say that publish_display_data is not connecting to the notebook. -Luke On 07/30/2014 05:41 PM, ipython-dev-request at scipy.org wrote: > ------------------------------ > > Message: 6 > Date: Wed, 30 Jul 2014 17:54:42 -0700 > From: Thomas Kluyver > Subject: Re: [IPython-dev] IDL/GDL kernel > To: IPython developers list > Message-ID: > > Content-Type: text/plain; charset="utf-8" > > Hi Luke, > > On 30 July 2014 17:29, Luke Stagner wrote: > >> >I am a graduate student studying plasma physics. I have made a mostly >> >functional IDL/GDL(just GDL for now) kernel >> > using the new wrapper kernel >> >feature. I borrowed heavily from the bash_kernel >> > example. If it is not too much >> >trouble I need some help with two things. >> > > Cool, I'm glad to hear that the API and the example are useful. > > >> > 1. IPython doesn't kill GDL process when shutting down or restarting the >> >kernel. I have defined the do_shutdown >> > >> >to kill the process but that doesn't seem to do anything. >> > > That's strange. I've just tried with my bash kernel, which doesn't > implement do_shutdown at all, and it does stop the process correctly (the > server stops the wrapper process, and that automatically causes the child > process to die). I don't know exactly what could be interfering with that, > but I'll keep thinking. > > >> >2. Inline Plots: My current method >> > is >> >to have GDL make pngs of the plots and then publishing them. Unfortunately >> >this is not working and I am not sure why. >> > > It should work. Looking at your code, you should switch back to the > commented out line that opens and reads the image files, rather than > displaying the filenames. If you comment out the rmtree(plot_dir) line, are > the images being written as you expect? > > Thomas > -------------- next part -------------- > An HTML attachment was scrubbed... > URL:http://mail.scipy.org/pipermail/ipython-dev/attachments/20140730/0ed4b32f/attachment.html > > ------------------------------ From takowl at gmail.com Wed Jul 30 22:30:18 2014 From: takowl at gmail.com (Thomas Kluyver) Date: Wed, 30 Jul 2014 19:30:18 -0700 Subject: [IPython-dev] IDL/GDL kernel In-Reply-To: <53D99C9D.4070900@uci.edu> References: <53D99C9D.4070900@uci.edu> Message-ID: On 30 July 2014 18:32, Luke Stagner wrote: > I have done what you have suggested and I verified that it is producing > the .png files. If I were to hazard a guess I would say that > publish_display_data is not connecting to the notebook. > Ah, digging into the code, you're right. publish_display_data depends on the IPython shell machinery, which is not being spun up for non-IPython kernels. You should be able to do it like this: self.send_response(self.iopub_socket, 'display_data', {'image/png': data}) Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From takowl at gmail.com Wed Jul 30 22:47:13 2014 From: takowl at gmail.com (Thomas Kluyver) Date: Wed, 30 Jul 2014 19:47:13 -0700 Subject: [IPython-dev] Public-domain browser terminal interface In-Reply-To: References:

Message-ID: The author of GraphTerm has just contacted me to say that he's pulled out the plain web terminal component - a combination of frontend stuff from term.js and updated Python code on the backend (so we don't have to handle more JS than we have to). It's BSD licensed, based on Tornado, uses websockets and appears to support Python 3, and Dr. Saravanan says he'd be happy to help us integrate it into IPython. https://github.com/mitotic/pyxterm Thomas On 8 July 2014 13:42, William Stein wrote: > On Tue, Jul 8, 2014 at 1:31 PM, Fernando Perez > wrote: > > > > On Tue, Jul 8, 2014 at 3:26 PM, William Stein wrote: > >> > >> What about term.js? It's MIT licensed (partly due to me asking > >> nicely), and is what many online IDE projects (including > >> SageMathCloud) use: > >> > >> https://github.com/chjj/term.js/ > > > > > > Thanks, that's the link I had in mind!! > > > >> > >> Maybe it isn't an option for you because the backend is Node.js > >> instead of Python. I just wanted to double check, just in case. > > > > > > Not a problem at all. We're thinking of this for the multiuser server, > which > > has a Node component as well, so it's perfect. > > > > We might take you up on your 'pile of code' offer later, but we'll let > you > > know if we have the bandwidth for it. To start with, we'll probably be OK > > with the plain vanilla code as-is. > > > > Thanks again! > > One particular thing I added was resizing the backend pty in response > to changing the terminal size. This involved a little bit of new > tricky node.js code. > It works in node 0.8.x, but didn't work _last time I tried_ with Node > 0.10.x, which is why all SageMathCloud projects are still running > 0.8.x. > > Anyway, if you guys go with term.js you'll inevitably run into the > same issue, so keep me posted. > > -- W/illiam > > > > > 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 > > > > > > -- > 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 > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ellisonbg at gmail.com Wed Jul 30 22:50:52 2014 From: ellisonbg at gmail.com (Brian Granger) Date: Wed, 30 Jul 2014 19:50:52 -0700 Subject: [IPython-dev] Public-domain browser terminal interface In-Reply-To: References:

Message-ID: Yes, that is really great! That will make it much easier to offer the terminal in contexts where node.js isn't in the backend. Fantastic! On Wed, Jul 30, 2014 at 7:47 PM, Thomas Kluyver wrote: > The author of GraphTerm has just contacted me to say that he's pulled out > the plain web terminal component - a combination of frontend stuff from > term.js and updated Python code on the backend (so we don't have to handle > more JS than we have to). It's BSD licensed, based on Tornado, uses > websockets and appears to support Python 3, and Dr. Saravanan says he'd be > happy to help us integrate it into IPython. > > https://github.com/mitotic/pyxterm > > Thomas > > > On 8 July 2014 13:42, William Stein wrote: >> >> On Tue, Jul 8, 2014 at 1:31 PM, Fernando Perez >> wrote: >> > >> > On Tue, Jul 8, 2014 at 3:26 PM, William Stein wrote: >> >> >> >> What about term.js? It's MIT licensed (partly due to me asking >> >> nicely), and is what many online IDE projects (including >> >> SageMathCloud) use: >> >> >> >> https://github.com/chjj/term.js/ >> > >> > >> > Thanks, that's the link I had in mind!! >> > >> >> >> >> Maybe it isn't an option for you because the backend is Node.js >> >> instead of Python. I just wanted to double check, just in case. >> > >> > >> > Not a problem at all. We're thinking of this for the multiuser server, >> > which >> > has a Node component as well, so it's perfect. >> > >> > We might take you up on your 'pile of code' offer later, but we'll let >> > you >> > know if we have the bandwidth for it. To start with, we'll probably be >> > OK >> > with the plain vanilla code as-is. >> > >> > Thanks again! >> >> One particular thing I added was resizing the backend pty in response >> to changing the terminal size. This involved a little bit of new >> tricky node.js code. >> It works in node 0.8.x, but didn't work _last time I tried_ with Node >> 0.10.x, which is why all SageMathCloud projects are still running >> 0.8.x. >> >> Anyway, if you guys go with term.js you'll inevitably run into the >> same issue, so keep me posted. >> >> -- W/illiam >> >> > >> > 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 >> > >> >> >> >> -- >> 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 > > > > _______________________________________________ > 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 @ellisonbg on Twitter and GitHub bgranger at calpoly.edu and ellisonbg at gmail.com From fperez.net at gmail.com Wed Jul 30 23:10:53 2014 From: fperez.net at gmail.com (Fernando Perez) Date: Wed, 30 Jul 2014 20:10:53 -0700 Subject: [IPython-dev] Public-domain browser terminal interface In-Reply-To: References:

Message-ID: On Wed, Jul 30, 2014 at 7:50 PM, Brian Granger wrote: > Yes, that is really great! That will make it much easier to offer the > terminal in contexts where node.js isn't in the backend. Fantastic! > +lots!! Very happy to hear this, thanks Thomas for following up... -- 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 lstagner at uci.edu Wed Jul 30 23:27:28 2014 From: lstagner at uci.edu (Luke Stagner) Date: Wed, 30 Jul 2014 20:27:28 -0700 Subject: [IPython-dev] IDL/GDL kernel In-Reply-To: References: <53D99C9D.4070900@uci.edu> Message-ID: <53D9B7A0.5050309@uci.edu> Hmm, that didn't seem to work. I now get the following error with the following revision . I must be doing something wrong. -Luke On 07/30/2014 07:30 PM, Thomas Kluyver wrote: > On 30 July 2014 18:32, Luke Stagner > wrote: > > I have done what you have suggested and I verified that it is > producing > the .png files. If I were to hazard a guess I would say that > publish_display_data is not connecting to the notebook. > > > Ah, digging into the code, you're right. publish_display_data depends > on the IPython shell machinery, which is not being spun up for > non-IPython kernels. > > You should be able to do it like this: > self.send_response(self.iopub_socket, 'display_data', {'image/png': data}) > > 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 Wed Jul 30 23:40:35 2014 From: takowl at gmail.com (Thomas Kluyver) Date: Wed, 30 Jul 2014 20:40:35 -0700 Subject: [IPython-dev] IDL/GDL kernel In-Reply-To: <53D9B7A0.5050309@uci.edu> References: <53D99C9D.4070900@uci.edu> <53D9B7A0.5050309@uci.edu> Message-ID: On 30 July 2014 20:27, Luke Stagner wrote: > Hmm, that didn't seem to work. I now get the following error > > with the following revision > . > I must be doing something wrong. > Oh yes, using that lower level interface, you'll need to base64 encode the data yourself. JSON can only handle text, not arbitrary bytes. Thomas -------------- next part -------------- An HTML attachment was scrubbed... URL: From lstagner at uci.edu Thu Jul 31 00:04:45 2014 From: lstagner at uci.edu (Luke Stagner) Date: Wed, 30 Jul 2014 21:04:45 -0700 Subject: [IPython-dev] IDL/GDL kernel In-Reply-To: References: <53D99C9D.4070900@uci.edu> <53D9B7A0.5050309@uci.edu> Message-ID: <53D9C05D.1030204@uci.edu> Good news no more error. Bad news still no inline plot. I even set it so that it reads some random png I have every time and still no go. Luke On 07/30/2014 08:40 PM, Thomas Kluyver wrote: > On 30 July 2014 20:27, Luke Stagner > wrote: > > Hmm, that didn't seem to work. I now get the following error > > with the following revision > . > I must be doing something wrong. > > > Oh yes, using that lower level interface, you'll need to base64 encode > the data yourself. JSON can only handle text, not arbitrary bytes. > > 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 lstagner at uci.edu Thu Jul 31 02:52:10 2014 From: lstagner at uci.edu (Luke Stagner) Date: Wed, 30 Jul 2014 23:52:10 -0700 Subject: [IPython-dev] IDL/GDL kernel In-Reply-To: References: <53D99C9D.4070900@uci.edu> <53D9B7A0.5050309@uci.edu> Message-ID: <53D9E79A.1070603@uci.edu> In regards to the kernel not shutting down the GDL process. It seems that kernel.do_shutdown is not being overridden by GDLKernel.do_shutdown. To check this I threw in an assert that will fail if GDLKernel.do_shutdown gets called. Suffice to say that assert is never thrown. -Luke On 07/30/2014 08:40 PM, Thomas Kluyver wrote: > On 30 July 2014 20:27, Luke Stagner > wrote: > > Hmm, that didn't seem to work. I now get the following error > > with the following revision > . > I must be doing something wrong. > > > Oh yes, using that lower level interface, you'll need to base64 encode > the data yourself. JSON can only handle text, not arbitrary bytes. > > 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 p.f.moore at gmail.com Thu Jul 31 03:00:38 2014 From: p.f.moore at gmail.com (Paul Moore) Date: Thu, 31 Jul 2014 08:00:38 +0100 Subject: [IPython-dev] IDL/GDL kernel In-Reply-To: <53D9E79A.1070603@uci.edu> References: <53D99C9D.4070900@uci.edu> <53D9B7A0.5050309@uci.edu> <53D9E79A.1070603@uci.edu> Message-ID: On 31 July 2014 07:52, Luke Stagner wrote: > In regards to the kernel not shutting down the GDL process. It seems that > kernel.do_shutdown is not being overridden by GDLKernel.do_shutdown. To > check this I threw in an assert that will fail if GDLKernel.do_shutdown gets > called. Suffice to say that assert is never thrown. >From visual inspection, it appears to me that do_shutdown is indented too far, making it a local function of do_execute, rather than a method of the class. Paul From lstagner at uci.edu Thu Jul 31 03:07:05 2014 From: lstagner at uci.edu (Luke Stagner) Date: Thu, 31 Jul 2014 00:07:05 -0700 Subject: [IPython-dev] IDL/GDL kernel In-Reply-To: References: <53D99C9D.4070900@uci.edu> <53D9B7A0.5050309@uci.edu> <53D9E79A.1070603@uci.edu> Message-ID: <53D9EB19.70104@uci.edu> I always suspected that I had no idea what I was doing. Its nice to finally have proof. On 07/31/2014 12:00 AM, Paul Moore wrote: > On 31 July 2014 07:52, Luke Stagner wrote: >> In regards to the kernel not shutting down the GDL process. It seems that >> kernel.do_shutdown is not being overridden by GDLKernel.do_shutdown. To >> check this I threw in an assert that will fail if GDLKernel.do_shutdown gets >> called. Suffice to say that assert is never thrown. > >From visual inspection, it appears to me that do_shutdown is indented > too far, making it a local function of do_execute, rather than a > method of the class. > > Paul > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > From ingolf.becker at googlemail.com Thu Jul 31 06:18:42 2014 From: ingolf.becker at googlemail.com (Ingolf Becker) Date: Thu, 31 Jul 2014 11:18:42 +0100 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing In-Reply-To: References: Message-ID: Dear Mark very nice notebooks! I particularl On 29 July 2014 21:56, Mark Bakker wrote: > We have developed a series of 15 Notebooks for scientists and engineers > who want to use Python programming for exploratory computing, scipting, > data analysis, and visualization.The Notebooks were developed for an > undergraduate class (sophomores) in Civil Engineering. No prior knowledge > of computer programming is assumed. Each Notebook covers a specific topic > and includes a number of exercises. The exercises should take less than 4 > hours to complete for each Notebook. > > The Notebooks may be viewed at > http://mbakker7.github.io/exploratory_computing_with_python/ > A link to the GitHub repository is also shown on this page. > > Hope you find them useful, > > Mark Bakker > TU Delft, The Netherlands > > _______________________________________________ > 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 ingolf.becker at googlemail.com Thu Jul 31 06:25:08 2014 From: ingolf.becker at googlemail.com (Ingolf Becker) Date: Thu, 31 Jul 2014 11:25:08 +0100 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing In-Reply-To: References: Message-ID: Sorry about that - I particularly like the notebooks on water management and the time series analysis. Just a remark though: I would avoid using %pylab inline - it has some hideous side effects [1] (and there was talk of it being removed entirely in future versions) `%matplotlib inline from matplotlib import pyplot as plt` should be preferred. [1] https://carreau.github.io/posts/10-No-PyLab-Thanks.ipynb.html Thank you for sharing! On 31 July 2014 11:18, Ingolf Becker wrote: > Dear Mark > > very nice notebooks! I particularl > > > On 29 July 2014 21:56, Mark Bakker wrote: > >> We have developed a series of 15 Notebooks for scientists and engineers >> who want to use Python programming for exploratory computing, scipting, >> data analysis, and visualization.The Notebooks were developed for an >> undergraduate class (sophomores) in Civil Engineering. No prior >> knowledge of computer programming is assumed. Each Notebook covers a >> specific topic and includes a number of exercises. The exercises should >> take less than 4 hours to complete for each Notebook. >> >> The Notebooks may be viewed at >> http://mbakker7.github.io/exploratory_computing_with_python/ >> A link to the GitHub repository is also shown on this page. >> >> Hope you find them useful, >> >> Mark Bakker >> TU Delft, The Netherlands >> >> _______________________________________________ >> 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 gelonida at gmail.com Thu Jul 31 07:30:31 2014 From: gelonida at gmail.com (Gelonida) Date: Thu, 31 Jul 2014 13:30:31 +0200 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing In-Reply-To: References: Message-ID: <53DA28D7.3040404@gmail.com> Hi Mark, Small suggestion to add some clarifying information to your link. On 7/29/2014 10:56 PM, Mark Bakker wrote: > We have developed a series of 15 Notebooks for scientists and > engineers who want to use Python programming for exploratory > computing, scipting, data analysis, and visualization.The Notebooks > were developed for an undergraduate class (sophomores) in Civil > Engineering. No prior knowledge of computer programming is assumed. > Each Notebook covers a specific topic and includes a number of > exercises. The exercises should take less than 4 hours to complete for > each Notebook. > > The Notebooks may be viewed at > http://mbakker7.github.io/exploratory_computing_with_python/ > A link to the GitHub repository is also shown on this page. > These links look interesting and I know a few persons who might enjoy learning a little about python this way. However as a notebook newbie (I only used ipython as interactive shell so far and did not read more about notebooks) I am a little confused. The web page seems to be aimed at newbies. It gives the impression, that I just have to go to http://mbakker7.github.io/exploratory_computing_with_python/ and then to click on notebook 1 ( http://nbviewer.ipython.org/github/mbakker7/exploratory_computing_with_python/blob/master/notebook1/py_exploratory_comp_1_sol.ipynb ) and that I could then change the contents of the input field, press shift-enter and then see the new result. But it seems, that this is just a static version of the ipython notebook. Only when - downloading the file - installing all package dependencies (IMHO very difficult for python newbies at least on windows) , - starting a cmd window - and running C:\Python27\Scripts\ipython notebook I was able to use the notebook as described. I assume, that with above procedure you'd loose about 99% of the newbies who could have been interested in trying out python/notebooks. Wouldn't it be possible to add a more explicit link which mentions at least, that the notebooks cannot be executed immediately or even better a link to a page trying to explain (even if not aimed for newbies) how to use/install ipython/notebooks? _ _ -------------- next part -------------- An HTML attachment was scrubbed... URL: From gvwilson at third-bit.com Thu Jul 31 13:56:47 2014 From: gvwilson at third-bit.com (Greg Wilson) Date: Thu, 31 Jul 2014 13:56:47 -0400 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing In-Reply-To: References:

Message-ID: <53DA835F.7050000@third-bit.com> Mark Bakker > wrote: We have developed a series of 15 Notebooks for scientists and engineers who want to use Python programming for exploratory computing, scipting, data analysis, and visualization.... Satrajit Ghosh wrote: > - how could we disseminate or find such modules that can be used for > teaching, without having to go through every single notebook? > - how do we describe the prior knowledge required through a dependency > structure? Greg Wilson writes: During our sprint last week, R?mi Emonet and Raniere Silva started exploring the idea of using off-the-shelf package managers to manage lesson material [1,2] so that something like: $ conda install barba-12-steps-navier-stokes will do what it should. I'm really excited by the idea, and hope to put some time into it once September rolls around - if people are interested, I'll send a note to this list once work gets under way. Thanks, Greg Wilson [1] https://softwarecarpentrylessonmanager.github.io/lesson-manager/04-howto.html [2] http://blog.rgaiacs.com/2014/07/27/sprint.html -------------- next part -------------- An HTML attachment was scrubbed... URL: From markbak at gmail.com Thu Jul 31 15:44:58 2014 From: markbak at gmail.com (Mark Bakker) Date: Thu, 31 Jul 2014 21:44:58 +0200 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing Message-ID: Yeah, good point about the %matplotlib inline. That is much better. The thing is that I am using Canopy in class and when you start an IPython Notebook in Canopy I think it starts by default in what used to be pylab mode. I wonder what it does now, as the --pylab flag is gone. Should try that, as this was the behavior in the class last year. From: Ingolf Becker > > I particularly like the notebooks on water management and the time series > analysis. Just a remark though: > I would avoid using %pylab inline - it has some hideous side effects [1] > (and there was talk of it being removed entirely in future versions) > `%matplotlib inline > from matplotlib import pyplot as plt` > should be preferred. > > [1] https://carreau.github.io/posts/10-No-PyLab-Thanks.ipynb.html > > Thank you for sharing! > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From markbak at gmail.com Thu Jul 31 15:49:53 2014 From: markbak at gmail.com (Mark Bakker) Date: Thu, 31 Jul 2014 21:49:53 +0200 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing Message-ID: Good point. For the class I obviously had installation instructions. They are quite easy when you install a packaged bundle (Canopy Express, PythonXY, Anaconda). We used Canopy Express in the class and that worked well. All Notebooks were also accompanied by 10-15 minute videos outlining the new features that were used in the Notebook. I will have to redo some of those videos and will add links to them in the next few weeks. For now I will add some information about installing and getting started. That is indeed a very good idea and should get newbies up and running, Mark On Thu, Jul 31, 2014 at 7:00 PM, wrote: > From: Gelonida > Hi Mark, > > > Small suggestion to add some clarifying information to your link. > > These links look interesting and I know a few persons who might enjoy > learning a little about python this way. > However as a notebook newbie (I only used ipython as interactive shell > so far and did not read more about notebooks) I am a little confused. > > The web page seems to be aimed at newbies. > It gives the impression, that I just have to go to > http://mbakker7.github.io/exploratory_computing_with_python/ > and then to click on notebook 1 ( > > http://nbviewer.ipython.org/github/mbakker7/exploratory_computing_with_python/blob/master/notebook1/py_exploratory_comp_1_sol.ipynb > ) > and that I could then change the contents of the input field, press > shift-enter and then see the new result. > > But it seems, that this is just a static version of the ipython notebook. > > Only when > - downloading the file > - installing all package dependencies (IMHO very difficult for python > newbies at least on windows) , > - starting a cmd window > - and running C:\Python27\Scripts\ipython notebook > > I was able to use the notebook as described. > > I assume, that with above procedure you'd loose about 99% of the newbies > who could have been interested in trying out python/notebooks. > > Wouldn't it be possible to add a more explicit link which mentions at > least, that the notebooks cannot be executed immediately or even better > a link to a page trying to explain (even if not aimed for newbies) > how to use/install ipython/notebooks? > -------------- next part -------------- An HTML attachment was scrubbed... URL: From takowl at gmail.com Thu Jul 31 18:55:01 2014 From: takowl at gmail.com (Thomas Kluyver) Date: Thu, 31 Jul 2014 15:55:01 -0700 Subject: [IPython-dev] IDL/GDL kernel In-Reply-To: <53D9C05D.1030204@uci.edu> References: <53D99C9D.4070900@uci.edu> <53D9B7A0.5050309@uci.edu> <53D9C05D.1030204@uci.edu> Message-ID: On 30 July 2014 21:04, Luke Stagner wrote: > Good news no more error. Bad news still no inline plot. I even set it so > that it reads some random png I have every time and still no go. > Sorry, I got the message structure wrong when I gave you an example without trying it. This time, here's an example that I have tested and ensured that it works: with open('foo.png', 'rb') as f: import base64 imgdata = base64.b64encode(f.read()) self.send_response(self.iopub_socket, 'display_data', {'data':{'image/png': imgdata.decode('ascii')}}) -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Thu Jul 31 19:01:29 2014 From: fperez.net at gmail.com (Fernando Perez) Date: Thu, 31 Jul 2014 16:01:29 -0700 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing In-Reply-To: <53DA835F.7050000@third-bit.com> References:

<53DA835F.7050000@third-bit.com> Message-ID: On Thu, Jul 31, 2014 at 10:56 AM, Greg Wilson wrote: > During our sprint last week, R?mi Emonet and Raniere Silva started > exploring the idea of using off-the-shelf package managers to manage lesson > material [1,2] so that something like: > > $ conda install barba-12-steps-navier-stokes > > will do what it should. I'm really excited by the idea, and hope to put > some time into it once September rolls around - if people are interested, > I'll send a note to this list once work gets under way. > Please do so! I've been thinking lately a lot along those lines as well, the 'code and data (up to some size limits) as a package'. I think there's a lot of potential in that line of thinking, and we may be able to leverage a lot of existing technology to make it easier to reuse bundles of material that focus on a specific teaching or research context, making both education and reproducible research easier. The fact that there are valid limits to the above with enormous data sizes shouldn't deter us from making progress on the many use cases where that won't be an issue. 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 Thu Jul 31 19:02:46 2014 From: fperez.net at gmail.com (Fernando Perez) Date: Thu, 31 Jul 2014 16:02:46 -0700 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing In-Reply-To: References: Message-ID: On Thu, Jul 31, 2014 at 12:44 PM, Mark Bakker wrote: > Yeah, good point about the %matplotlib inline. That is much better. The > thing is that I am using Canopy in class and when you start an IPython > Notebook in Canopy I think it starts by default in what used to be pylab > mode. I wonder what it does now, as the --pylab flag is gone. Should try > that, as this was the behavior in the class last year. Just ignore that: if you add the explicit %matplotlib call and the manual imports, the code will still work in canopy, and it will *also* work for anyone who starts without the global --pylab flag. So just do the 'new right thing' manually with %matplotlib, and it will work fine transparently. 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 Thu Jul 31 19:08:14 2014 From: fperez.net at gmail.com (Fernando Perez) Date: Thu, 31 Jul 2014 16:08:14 -0700 Subject: [IPython-dev] IDL/GDL kernel In-Reply-To: References: <53D99C9D.4070900@uci.edu> <53D9B7A0.5050309@uci.edu> <53D9C05D.1030204@uci.edu> Message-ID: Do you think it's worth adding this to the base class as a utility method(s)? On Thu, Jul 31, 2014 at 3:55 PM, Thomas Kluyver wrote: > On 30 July 2014 21:04, Luke Stagner wrote: > >> Good news no more error. Bad news still no inline plot. I even set it so >> that it reads some random png I have every time and still no go. >> > > Sorry, I got the message structure wrong when I gave you an example > without trying it. This time, here's an example that I have tested and > ensured that it works: > > with open('foo.png', 'rb') as f: > > import base64 > > imgdata = base64.b64encode(f.read()) > > self.send_response(self.iopub_socket, 'display_data', > {'data':{'image/png': imgdata.decode('ascii')}}) > > > _______________________________________________ > 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 doug.blank at gmail.com Thu Jul 31 20:36:51 2014 From: doug.blank at gmail.com (Doug Blank) Date: Thu, 31 Jul 2014 20:36:51 -0400 Subject: [IPython-dev] Install and Demo Videos of Notebook Extensions Message-ID: Here are some new videos showing installation and demonstration of the Spell Check, Document Tools, and Cell Tools. Document Tools includes section movers, heading numbering, table of contents, and bibliographic tools (biblio not demoed yet) Cell Tools includes Input/Output tabs, and two column view. Installation videos: 1) Jupyter Notebook Cell Tools Extension Installation (Windows) https://www.youtube.com/watch?v=WwoTzvOkEJQ 2) Jupyter Notebook Document Tools Extension Installation (Windows) https://www.youtube.com/watch?v=YbM8rrj-Bms 3) Jupyter Notebook Spell Check Extension Installation (Windows) https://www.youtube.com/watch?v=Km3AtRynWFQ Demonstration videos: 4) Jupyter Notebook Cell Tools Extension Demonstration https://www.youtube.com/watch?v=dBB1TCWQgBg 5) Jupyter Notebook Document Tools Extension Demonstration https://www.youtube.com/watch?v=LRzUSilYEks 6) Jupyter Notebook Spell Check Extension Demonstration https://www.youtube.com/watch?v=o4xCp3b4oCw -Doug -------------- next part -------------- An HTML attachment was scrubbed... URL: From satra at mit.edu Thu Jul 31 20:58:27 2014 From: satra at mit.edu (Satrajit Ghosh) Date: Thu, 31 Jul 2014 20:58:27 -0400 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing In-Reply-To: References:

<53DA835F.7050000@third-bit.com> Message-ID: hi greg and fernando, along these both git-annex [1] and ipfs [2] might be solutions for the near future, and would offer easy flexibility across data sizes. cheers, satra [1] https://git-annex.branchable.com/ [2] http://ipfs.io On Thu, Jul 31, 2014 at 7:01 PM, Fernando Perez wrote: > > On Thu, Jul 31, 2014 at 10:56 AM, Greg Wilson > wrote: > >> During our sprint last week, R?mi Emonet and Raniere Silva started >> exploring the idea of using off-the-shelf package managers to manage lesson >> material [1,2] so that something like: >> >> $ conda install barba-12-steps-navier-stokes >> >> will do what it should. I'm really excited by the idea, and hope to put >> some time into it once September rolls around - if people are interested, >> I'll send a note to this list once work gets under way. >> > > Please do so! I've been thinking lately a lot along those lines as well, > the 'code and data (up to some size limits) as a package'. I think there's > a lot of potential in that line of thinking, and we may be able to leverage > a lot of existing technology to make it easier to reuse bundles of material > that focus on a specific teaching or research context, making both > education and reproducible research easier. > > The fact that there are valid limits to the above with enormous data sizes > shouldn't deter us from making progress on the many use cases where that > won't be an issue. > > 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 nathan12343 at gmail.com Thu Jul 31 21:11:30 2014 From: nathan12343 at gmail.com (Nathan Goldbaum) Date: Thu, 31 Jul 2014 18:11:30 -0700 Subject: [IPython-dev] Install and Demo Videos of Notebook Extensions In-Reply-To: References: Message-ID: Thanks very much for forwarding the videos - it was very nice to have the demo to see what your extensions look like in practice. I believe this was already discussed in an earlier thread, but it would be very nice if there were a way to select suggestions for misspelled words. I do this all time time using chrome's spell-checker, so it's a bit of a habit to right-click on words that have squiggly red underlines. I guess some people don't like messing with the right-click menu, so maybe it could be toggled somehow? In any case, having indicators about misspelled words at all is a huge improvement. Thanks for sharing the extension. -Nathan On Thu, Jul 31, 2014 at 5:36 PM, Doug Blank wrote: > Here are some new videos showing installation and demonstration of the > Spell Check, Document Tools, and Cell Tools. > > Document Tools includes section movers, heading numbering, table of > contents, and bibliographic tools (biblio not demoed yet) > > Cell Tools includes Input/Output tabs, and two column view. > > Installation videos: > > 1) Jupyter Notebook Cell Tools Extension Installation (Windows) > https://www.youtube.com/watch?v=WwoTzvOkEJQ > > 2) Jupyter Notebook Document Tools Extension Installation (Windows) > https://www.youtube.com/watch?v=YbM8rrj-Bms > > 3) Jupyter Notebook Spell Check Extension Installation (Windows) > https://www.youtube.com/watch?v=Km3AtRynWFQ > > Demonstration videos: > > 4) Jupyter Notebook Cell Tools Extension Demonstration > https://www.youtube.com/watch?v=dBB1TCWQgBg > > 5) Jupyter Notebook Document Tools Extension Demonstration > https://www.youtube.com/watch?v=LRzUSilYEks > > 6) Jupyter Notebook Spell Check Extension Demonstration > https://www.youtube.com/watch?v=o4xCp3b4oCw > > -Doug > > > _______________________________________________ > 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 gvwilson at third-bit.com Thu Jul 31 21:39:47 2014 From: gvwilson at third-bit.com (Greg Wilson) Date: Thu, 31 Jul 2014 21:39:47 -0400 Subject: [IPython-dev] Announcement: Set of IPython Notebooks to learn Python for Exploratory Computing In-Reply-To: References:

<53DA835F.7050000@third-bit.com> Message-ID: <53DAEFE3.3000005@third-bit.com> On Thu, Jul 31, 2014 at 10:56 AM, Greg Wilson > wrote: During our sprint last week, R?mi Emonet and Raniere Silva started exploring the idea of using off-the-shelf package managers to manage lesson material [1,2] so that something like: $ conda install barba-12-steps-navier-stokes will do what it should. I'm really excited by the idea, and hope to put some time into it once September rolls around - if people are interested, I'll send a note to this list once work gets under way. On 2014-07-31 7:01 PM, Fernando Perez wrote: > Please do so! I've been thinking lately a lot along those lines as > well, the 'code and data (up to some size limits) as a package'. I > think there's a lot of potential in that line of thinking, and we may > be able to leverage a lot of existing technology to make it easier to > reuse bundles of material that focus on a specific teaching or > research context, making both education and reproducible research easier. Yeah - what I *actually* want to be able to do is: $ conda install doi://10.1371/journal.pone.0103452 and get the source of the paper, the code, the libraries the code depends on (in a safely sandboxed environment), and either the data or links to the data (or torrent files for the data). Lessons are then just a special case... Thanks, Greg -------------- next part -------------- An HTML attachment was scrubbed... URL: From doug.blank at gmail.com Thu Jul 31 21:46:16 2014 From: doug.blank at gmail.com (Doug Blank) Date: Thu, 31 Jul 2014 21:46:16 -0400 Subject: [IPython-dev] Install and Demo Videos of Notebook Extensions In-Reply-To: References: Message-ID: On Thu, Jul 31, 2014 at 9:11 PM, Nathan Goldbaum wrote: > Thanks very much for forwarding the videos - it was very nice to have the > demo to see what your extensions look like in practice. > You are welcomed, and I agree! Sometimes it is hard to see what an extension really is, and there isn't currently an uninstall. > > I believe this was already discussed in an earlier thread, but it would be > very nice if there were a way to select suggestions for misspelled words. > I do this all time time using chrome's spell-checker, so it's a bit of a > habit to right-click on words that have squiggly red underlines. I guess > some people don't like messing with the right-click menu, so maybe it could > be toggled somehow? > > In any case, having indicators about misspelled words at all is a huge > improvement. Thanks for sharing the extension. > I agree that we should eventually add that functionality. I'm going to be using the notebooks to teach writing this Fall. I'll explain that this current method (no suggestions) fits the philosophy of the system... it doesn't do everything for you, but tries to help along the way. And, just perhaps, students might actually learn how to spell :) But yes, in the long run, it should make suggestions on right-click. Someone in the issue tracker even suggested that we could hook up functionality to do tab word completion... that would be even more in the spirit of the system :) Might as well take advantage of the dictionary, if you've got it. -Doug > > -Nathan > > > On Thu, Jul 31, 2014 at 5:36 PM, Doug Blank wrote: > >> Here are some new videos showing installation and demonstration of the >> Spell Check, Document Tools, and Cell Tools. >> >> Document Tools includes section movers, heading numbering, table of >> contents, and bibliographic tools (biblio not demoed yet) >> >> Cell Tools includes Input/Output tabs, and two column view. >> >> Installation videos: >> >> 1) Jupyter Notebook Cell Tools Extension Installation (Windows) >> https://www.youtube.com/watch?v=WwoTzvOkEJQ >> >> 2) Jupyter Notebook Document Tools Extension Installation (Windows) >> https://www.youtube.com/watch?v=YbM8rrj-Bms >> >> 3) Jupyter Notebook Spell Check Extension Installation (Windows) >> https://www.youtube.com/watch?v=Km3AtRynWFQ >> >> Demonstration videos: >> >> 4) Jupyter Notebook Cell Tools Extension Demonstration >> https://www.youtube.com/watch?v=dBB1TCWQgBg >> >> 5) Jupyter Notebook Document Tools Extension Demonstration >> https://www.youtube.com/watch?v=LRzUSilYEks >> >> 6) Jupyter Notebook Spell Check Extension Demonstration >> https://www.youtube.com/watch?v=o4xCp3b4oCw >> >> -Doug >> >> >> _______________________________________________ >> 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: