These are chat archives for ipython/ipython

3rd
Mar 2016
bicubic
@bicubic
Mar 03 2016 05:12
Hrm. the pre_run_cell and pre_execute event is triggered but neither post_run_cell nor post_execute are triggered if there's a syntax error in a code cell. Is this by design?
If it is, is there an event that gets fired after cell processing is finished, whether syntax error or not?
sumit sinha
@sumitsinha91
Mar 03 2016 12:33
what is ipython
i am new on this
i am start working on django
i want to use ipython can anyone tell me if i will install ipython in my system then what i will get
can anyone give me basic idea
i needed
Fred E
@Fredilly
Mar 03 2016 12:36
Good question. It's a distribution environment for running data science code in python. Google "Anaconda ipython"
sumit sinha
@sumitsinha91
Mar 03 2016 12:37
hey i want to install
can you help me
how i will do it
Fred E
@Fredilly
Mar 03 2016 12:39
Sure, it's pretty straight forward. What OS do you run?
sumit sinha
@sumitsinha91
Mar 03 2016 12:39
ubuntu
64 bit
15.10 version
can you give me some basic command
Fred E
@Fredilly
Mar 03 2016 12:42
Fred E
@Fredilly
Mar 03 2016 12:47
Hope that helps. Let me know.
sumit sinha
@sumitsinha91
Mar 03 2016 12:47
sure let me check all this link
what is anconda python and what i will get from this
i am using django on my system
Fred E
@Fredilly
Mar 03 2016 12:53
You will get a python environment that has many useful libraries like pandas and numpy.
sumit sinha
@sumitsinha91
Mar 03 2016 12:54
so which type of help i will get from this
Fred E
@Fredilly
Mar 03 2016 12:57
You’ll have a complete environment for running your code in python and displaying results.
sumit sinha
@sumitsinha91
Mar 03 2016 12:57
ok
thanks
i want to expore more on python so can you help me
i want to learn more python django so from where i will more on this
Fred E
@Fredilly
Mar 03 2016 12:59
You can learn python at www.codeacademy.com
sumit sinha
@sumitsinha91
Mar 03 2016 12:59
i have completed from here
anyother you have website
that from where i will learn
Min RK
@minrk
Mar 03 2016 14:03
Thanks, @Fredilly!
Steven Silvester
Mar 03 2016 14:29
@minrk, @jasongrout, I’ve been playing with mathjax-node this morning, and am very impressed with it. The main difference from a consumption standpoint is that one must call typeset() manually for inline and block rendering. Any objections to using this in JupyterLab?
Min RK
@minrk
Mar 03 2016 14:36
The only objection I can think of is that two MathJax installs in the source is a bit excessive. I don't suppose there is a way to load the npm-installed one in the main app, too?
Steven Silvester
Mar 03 2016 14:36
oh, I see, from a total size perspective
yeah, we can surely replace mathjaxutils.js with a node version
(which is essentially what I’m doing now)
so, I’ll add the functionality in lab and make a PR to backport it to the existing noteboook
that’s a pretty equation I made this morning ;)
Steven Silvester
Mar 03 2016 14:54
interesting notes @minrk @jasongrout: the mathjax package on npm clocks in at 58M, mathjax-node is 252K, and katex is 424K
but of course mathjax-node depends on the mathjax package
Steven Silvester
Mar 03 2016 15:15
if we bundle the following, nothing needs to change otherwise:
import 'mathjax';
import 'mathjax/extensions/MathZoom’;
Min RK
@minrk
Mar 03 2016 15:24
Yeah, not an issue with conflict. Full MathJax is just so big. We excluded it for that reason for some time, but the bower package and some slimming made it acceptable.
We should see if slimming the node package is okay, too.
The vast majority of the size is image fonts that we don't use.
The next level of slimming is font selection and output formats.
In 4.1 we strip output formats, but we are rolling that back for 4.2 since it doesn't cost much. We do exclude all but one font, though.
Steven Silvester
Mar 03 2016 15:29
The webpack bundle with the above included is 1.7M
I’m still playing with it
I've just reopened a six-month-old notebook. Cells with %matplotlib nbagg don't render correctly. Googling I find %matplotlib notebook is now preferred. Doing that, I get "Warning: Cannot change to a different GUI toolkit: notebook. Using nbagg instead." But it does actually render okay!
Steven Silvester
Mar 03 2016 15:54
once you re-run the notebook, that warning should go away
(or re-run that cell)
Thanks. Also, how do I keep my anaconda up-to-date?
Steven Silvester
Mar 03 2016 15:56
conda update conda; conda update anaconda
great! btw you could integrate from -inf to +inf and get rid of the /2. Interesting result, I'm trying to figure out how one would derive it. I suspect using complex analysis.
Steven Silvester
Mar 03 2016 15:59
ha, thanks, that was just an example latex equation I was trying to render ;)
ouch http://hastebin.com/humoyileqe.coffee that doesn't look right!
Steven Silvester
Mar 03 2016 16:03
ah, yes, the main anaconda enviroment is 3.5 now
you want conda update —all
tx
Steven Silvester
Mar 03 2016 16:10
no problemo
Jason Grout
@jasongrout
Mar 03 2016 17:49
@blink1073 - so what's the final size with the mathjax npm package in the notebook? How much does it impact the bottom line of code size?
I see 58M, but also 1.7M bundled...
in the conversation above
nice equation, by the way.
Steven Silvester