[jupyter] Re: How can I run cells not in sequence? (I'm asking for how it is implemented)

You might want to have a look at qtconsole as well, for additional inspiration: https://github.com/jupyter/qtconsole -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To unsubscribe from this group and stop receiving emails from it, send an em

[jupyter] Re: How to add custom syntax highlighting in ipython for Jupyter Lab

JupyterLab uses CodeMirror for syntax highlighting: https://github.com/jupyterlab/jupyterlab/issues/5504 -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To unsubscribe from this group and stop receiving emails from it, send an email to jupyt

[jupyter] Re: How can I run cells not in sequence? (I'm asking for how it is implemented)

When you execute cells in the Web UI, the browser sends the code of the cells to be executed to the notebook server. The server itself doesn't know about cells, it just gets code snippets for execution. Your alternative implementation of a console-based UI will have to work in a similar fashion

[jupyter] Re: Sharing kernels along with personal kernels

Hello Sean, there's an environment variable JUPYTER_PATH that you can populate with additional directories to be searched. https://jupyter.readthedocs.io/en/latest/projects/jupyter-directories.html#data-files Cheers, Roland -- You received this message because you are subscribed to the Googl

Re: [jupyter] How does Kernel Maintain context of previous cells in a session

On Wednesday, September 11, 2019 at 5:57:46 PM UTC+2, zweb1 wrote: > > I assume namespace that gets passed around only stores the named reference > to objects and not the data itself. > > Data and objects are in memory separately so it should be only limited by > memory of server or is it limite

Re: [jupyter] Re: Issue: Jupyter Notebook

Yes, sounds more like an Apache Drill problem. In addition to the Apache Drill community, you can also try here or on discourse.jupyter.org while mentioning "Apache Drill" in the subject. It#s all about getting the right people's attention :-) cheers, Roland -- You received this message bec

[jupyter] Re: Issue: Jupyter Notebook

Is this an error that you get when you try to run Jupyter? Or is Jupyter running fine, and you get the error when you try to use Apache Drill from within a Jupyter session? In the latter case, the subject you have chosen would be quite misleading. The jupyter_drill readme has a few lines of cod

[jupyter] Re: jupyter nbconvert - recursive option request?

Personally, I don't think that nbconvert should bother with searching input files in a directory structure. As your post shows, this is easily added through scripting, without causing extra effort for the nbconvert maintainers. Nevertheless, you can always open a feature request: https://github

[jupyter] Re: permissions problem in jupyter on windows 10

Hello, posting the failing code and the error message directly will improve your chances of getting a useful response. I'm probably not the only one who hates looking at .ipynb files in raw format. I did not find a permission error in your file. There's an error messages about an 'mpg.csv' fil

[jupyter] Re: No such file or directory

What do you mean with I set the path of the data folder. > I found an "os.path.join" in the first .ipynb I looked at. It seems to construct a path that is relative to the working directory, not to a data directory. As a side note, pointing folks at a whole GitHub repo instead of putting the r

[jupyter] Re: Possibility to separate the JupyterLab Web and Server

On Sunday, April 28, 2019 at 8:23:13 PM UTC+2, Song Liu wrote: > > So that actually I am looking for a deployment separation between "static > (html,js)" and the backend server, where: > > - the web server provides the viewing and editing (+ save to storage) > function > - the backend server is

[jupyter] Re: How to download multiple files in Jupyter?

If you're using Python, call !tar or !zip to create a single archive that contains your whole folder. For other languages, there are system calls to execute shell commands with the same effect. cheers, Roland PS: The fastest and the only certain way to get a feature request implemented is t

[jupyter] Re: How could I change default language settings in Jupyter notebook?

It might be your browser requesting the Chinese language from Jupyter. In that case, you'll have to change browser language settings. See this discussion here: https://groups.google.com/d/topic/jupyter/6VA4nA_6jZ0/discussion cheers, Roland -- You received this message because you are subscri

[jupyter] Re: Assistance required starting jupyter-notbook through Anaconda-Navigator on linux

Thanks for the root cause analysis ;-) Another solution might be to run anaconda-navigator and thereby Jupyter as non-root. cheers, Roland -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To unsubscribe from this group and stop receiving

[jupyter] Re: Assistance required starting jupyter-notbook through Anaconda-Navigator on linux

To get the problem with Anaconda Navigator sorted out, you'll have to reach out to Anaconda. The first error message suggests that it expects but doesn't find (a certain version of) Visual Studio... which makes little sense on Linux. For Jupyter to work, you don't need Anaconda Navigator though

Re: [jupyter] stop executing cells programatically

Use the ternary operator as the last statement in the cell... data if TESTING else None -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To unsubscribe from this group and stop receiving emails from it, send an email to jupyter+unsubscr...@g

[jupyter] Re: REST API command to get the individual user notebook details - Requesting for help

Hello Praveen, each user is running their own notebook server(s). You'll have to connect to each of these servers in turn to query notebook status. But I don't think your admin token will be accepted by those notebook servers. You could change the authentication logic of the notebook servers to

[jupyter] Re: Requesting description for last activity response parameter.

Hello Praveen, I'd say one is the last activity of the user in general, the other is the last activity of that particular server of the user. The REST API looks as if a user can start multiple servers: https://jupyterhub.readthedocs.io/en/stable/_static/rest-api/index.html So the existing serve

[jupyter] Re: User Processes not releasing Memory

Those are background threads of libzmq, as mentioned here: https://github.com/zeromq/libzmq/blob/master/NEWS Jupyter uses ZMQ to connect to the running kernels on the local machine. There's not much to be done for the threads besides passing messages, so they won't accumulate a lot of CPU time.

[jupyter] Re: How to make automatic shutdown of notebook when notebooks are idle for long hours.

Have you checked that you're on a recent version of Jupyter Notebook? Are the kernels actually idle, or could they still be executing a long-running cell? Have you tried to enable debug logging, to see if the culling is actually attempted? -- You received this message because you are subscrib

[jupyter] Re: "To access the notebook, open this file in a browser:"

Hello Ted, you probably updated Jupyter Notebook. The notebook is no longer opened directly with a URL, but instead through a local .html file that redirects to the URL. That way, the URL with the token doesn't show in the process information anymore. Try setting the NotebookApp.browser config

[jupyter] Re: Working Javascript in HTML rendered notebooks

Hello Jason, Is there a way to ensure the javascript components work as per the actual > notebooks when the ipynb file has been converted to html? > In general? No. JavaScript that relies on data stored in the notebook might work. JavaScript that tries to interact with the running kernel will f

Re: [jupyter] Bug (feature?) in notebook >= 5.7

IIRC there was a recent change that the browser is no longer launched with a http(s): URL that points to the Jupyter server, but instead with a file: URL that points to a local HTML file containing a redirect to the http(s): URL. Maybe this change somehow interacts or interferes with browser la

[jupyter] Re: Jupyter uses wrong browser

Are you running "jupyter notebook" from the same environment in which you did the test with the webbrowser module? If it is a different environment (conda?), compare the environment variable BROWSER in both. cheers, Roland -- You received this message because you are subscribed to the Google

[jupyter] Re: pandoc 2.2.1 too new for nbconvert?

That restriction looks like a generic safeguard based on semantic versioning. With semantic versioning, a new major version indicates an incompatible API change. So somebody will have to look into it and adapt to the breaking changes, or at least verify that everything still works. I guess that

[jupyter] Re: Access local files from jupyter/scipy-notebook

If you mean data files, just use an absolute path to the file location. But the user ID which is running the kernel will need to have access to that path and file. cheers, Roland -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To unsubsc

Re: [jupyter] Re: Jupyter not running 'source' after opening from Anaconda exe

Hello Matt, you have something mixed up. The nb_conda module runs inside the Jupyter process and starts kernels in separate processes. It's the task of that module to establish the environment for those kernel processes. But it has no control over the environment of the Jupyter process itself.

[jupyter] Re: Jupyter not running 'source' after opening from Anaconda exe

Hello Matt, it is the nature of an "environment", including all environment variables, that it must be established _before_ invoking a command. Commands run within the environment in which you start them. They can respond to environment settings, or report errors if they cannot work within the

[jupyter] Re: spark in jupyter cannot find a class in a jar

That sounds like a problem between Py4J and Hadoop, or maybe pyspark. There's not a single appearance of anything from Jupyter in either the code or the error message you posted. I doubt that you will find much help for that problem in a Jupyter forum. Have you reached out to the Hadoop and/or

[jupyter] Re: Is pandas.types.common deprecated or not?

You're dealing with two different versions of Pandas. Check the version and install location in each environment like this: import pandas print(pandas.__version__) print(pandas.__file__) Before using the command prompt, are you activating the conda environment in which the kernel is

[jupyter] Re: jupyter-scipy:latest image cant work on centos6

Hello reminia, And it says "FATAL: kernel too old". > > Does this means the image only supports centos7 with higher docker version? > The error is not complaining about your Docker version, it is complaining about your Linux kernel version. You will need to upgrade your operating system. Of cou

[jupyter] Re: Pyspark Jupyter notebook not connected... but it IS

On Wednesday, October 24, 2018 at 8:20:27 AM UTC+2, kjell...@sony.com wrote: > > a progress bar for a heavy task keeps updating > ... > So why do the header then shows "Not connected" > In our environments, we've seen problems when a "heavy task" in a kernel is taking so much CPU that the Jupyter

Re: [jupyter] Re: Setting kernel values through request handlers

Hello Claude, For the auth, to begin with I was going to check on the logged_in status on > the IPythonHandler > The external server will not be logged in. Only the client browser is. You'll need to pass some authentication information from the client browser to the external server, so that th

[jupyter] Re: Setting kernel values through request handlers

Hello Claude, the notebook server and the kernel where your code cells execute are two different processes. The notebook server cannot directly update the variables in your kernel. You could translate the variable definitions received in your POST request to code, and send that code for execut

[jupyter] Re: Check the version of seaborn & sns.lineplot()

import seaborn print(seaborn.__version__) Jupyter Notebook is not responsible for the package versions in your kernel environment. You can install there whatever version of seaborn you want to use. If you want to upgrade, you'll also have to do that on your own. Depending on how you installed i

[jupyter] Re: Jupyter running under virtualenv is running "external" Python for notebook kernels

> > I'm still interested to understand why my virtualenv/cygwin/jupyter was > picking up my Anaconda Python3 kernel though - pour mourir moins bete as > they say around here. > If I read correctly, you started Jupyter with a plain 'jupyter' command, no absolute path. Have you check the PATH en

[jupyter] Re: How to point my Jupyter Notebook to miniconda Python installation?

I suggest to use these installation steps instead: a) install miniconda b) install Jupyter Notebook with miniconda Anaconda does a great job with packaging and integration. If you want to use conda anyway, it is best to do as much as possible with it. Unless your requirements are very special, t

[jupyter] Re: JupyterHub cannot talk to LDAP server

Hello Raymond, in the link you posted, the author decided to ignore the server certificate. While this is a common "solution" for accepting non-default certificates, it is also insecure. On a high level, you need to tell the TLS/SSL logic to trust the certificate which the administrator gave y

[jupyter] Re: Shell validated kinit/keytab gets "Invalid Syntax" error

Your notebook expects Python code, but you're giving it a shell command. While it is possible to execute shell commands in a Python notebook with the ! prefix, these commands will actually be executed in a sub-process, which terminates as soon as the command is executed. You cannot change key b

[jupyter] Re: Problem using another program.

Hello Bruno, it is not clear to me how your question is related to Jupyter. Are you trying to run your Python program from a notebook? If yes, which kernel language? From the top of my head, I don't know which language might provide a 'run' command. >From a Unix command line, you could start a

[jupyter] Re: Anaconda3 is installed, jupyter notebook errors out No module named 'pyspark'

Hello Raymond, how did you start Jupyter? The options in your .bashrc indicate that Anaconda noticed your Spark installation and prepared for starting jupyter through pyspark. But if you start Jupyter directly with plain Python, it won't know about Spark. The problem isn't with the code in your

[jupyter] Re: Introduction

Hello Rupjit, thanks for your kind offer. There are many ways in which you may participate in the Jupyter community: answer questions on the mailing list, write blogs and demos, improve documentation, fix bugs, implement features, and probably some more I've forgotten. The Contributor Guide se

[jupyter] Re: jupyterlab not honouring PYTHONPATH env var

Hello Rob, have a look at these discussions, maybe there's something that works for you: https://groups.google.com/d/topic/jupyter/yfg3O4EB7C0/discussion https://groups.google.com/d/topic/jupyter/UvwJ3NXcYZo/discussion cheers, Roland -- You received this message because you are subscribed to

[jupyter] Re: Dynamically change relative link extension from IPYNB to HTML when downloading as notebook as HTML?

How about giving the user a choice? File 2: [html](File2.html) [ipynb](File2.ipynb) -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To unsubscribe from this group and stop receiving emails from it, send an email to jupyter+unsubscr...@googl

[jupyter] Re: $PYTHONPATH for jupyter notebook (and jupyter notebook within jupyter lab)

You can add environment variables to the kernelspecs. -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To unsubscribe from this group and stop receiving emails from it, send an email to jupyter+unsubscr...@googlegroups.com. To post to this gr

[jupyter] Re: Running the proxy without SSL

Afaik, listening on two different ports is not possible without changing code. You could also solve your problem by keeping JupyterHub on the deprecated SSL port, and encrypting the connection between the SSL-terminating proxy and JupyterHub. I'm sure that pull requests to improve the document

[jupyter] Re: Reverting to a previous checkpoint

Restore from your backup? The default implementation keeps only one checkpoint. Unless you plugged in a different checkpointing implementation, there's no going back beyond one. best regards, Roland -- You received this message because you are subscribed to the Google Groups "Project Jupyte

[jupyter] Re: Environment variables PATH and PYTHONPATH not available in notebook

Have you tried setting BASH_ENV? https://www.gnu.org/software/bash/manual/html_node/Bash-Startup-Files.html /etc/profile should have worked in combination with --login only. .bashrc never had a chance, because the shell isn't interactive. -- You received this message because you are subscribed

Re: [jupyter] Jupyter Lab and ISO 8601

> > Not sure how I feel about showing users ISO 8601 formatted datetimes >> though. >> > It is not a particularly human friendly datetime format. >> > > It’s standard in Japan. And it *is* an international standard, after all. > Besides which, it makes logical sense. > Date and time formatting

Re: [jupyter] Blocking copy/paste functionality to JupyterHub

Hello Jeremy. Blocking copy and paste is one of the main features that we wish to > achieve. Essentially, we would like to ensure that a JupyterHub notebook > user is able to use and modify their notebooks at will normally (via > typing, clicking and normal keystrokes), > When I write code, I

[jupyter] Re: Updating Anaconda has broken Jupyter

On Thursday, February 22, 2018 at 1:48:51 AM UTC+1, gwhe...@ebay.com wrote: > > I ended up having to reinstall Anaconda. Not sure how it got so messed up > but it turned into a game of whack-a-mole and even creating new virtualenvs > didn't help. > Mixing conda envs with virtualenvs sounds like

[jupyter] Re: Converting python log to Notebook

Sounds like nbconvert is what you're looking for... https://nbconvert.readthedocs.io/en/latest/ -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To unsubscribe from this group and stop receiving emails from it, send an email to jupyter+unsubs

[jupyter] Re: Register external IPython kernel with jupyter notebook?

Hi Edward, I'm not going to read up on everything you posted here :-) I know the PID of the running kernel process. If I could somehow do a > subprocess.Popen() on that existing process and return a reference to that, > I think that would work. > Try starting a subprocess that joins or polls o

[jupyter] Re: How to know which config file is being used?

Hello Nate, which config file is being used by my jupyter folder? > Config files are used by processes, not by folders. Have you tried running "jupyter --paths" from a Python notebook? Because environment variables are inherited by default, this should show the locations where Jupyter looks fo

[jupyter] Re: Register external IPython kernel with jupyter notebook?

Hello Edward, through either making a custom server extension > > or using the jupyter kernel gateway > . > The kernel gateway in its websocket personality doesn't work with no

Re: [jupyter] Re: how come cells do not display all my code?

Hello Andy, maybe there's something in the default stylesheet. I know that long output is shown with a scrollbar by default, but you can click to the left of the output to have that section expanded. Could be that the same thing is in effect for code cells as well. cheers, Roland -- You re

[jupyter] Re: how come cells do not display all my code?

Hello Andy, is this about displaying the Python *code*, or the *output* when you run the code? Restarting the kernel should have no effect on what is being displayed. Unless you select "Restart & Clear Output", in which case it is expected that all output gets cleared. If your problem is miss

[jupyter] Re: Running Jupyter kernel gateway and notebook simultaneously

Hello Olav, as Luciano already pointed out, we need more information about your problem. Have you looked at the logs or console output of both notebook and kernel gateway? Is one of them coming up, or are both running into problems? Luciano, Olav is running the kernel gateway in notebook-http mo

Re: [jupyter] bug with logout?

On Saturday, January 20, 2018 at 12:08:16 AM UTC+1, Lawrence D’Oliveiro wrote: > > Surely it’s the other way round, the usual practice being to maintain a > store of *valid* tokens, with a finite lifetime attached to each (perhaps > reset when they get presented again). The tokens get deleted ei

[jupyter] Re: Resume training of neural network in the cloud using jupyter notebooks

Yes, you probably have to train again. The kernels don't know about notebooks. They send their output on channels, where it is received by the browser, and stored into the notebook you are working with. If the browser doesn't receive the output, it will not be stored. If your kernel is still r

[jupyter] Re: environment variables in terminal

As explained by Lawrence: set your LD_LIBRARY_PATH in .bashrc instead of .profile. -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To unsubscribe from this group and stop receiving emails from it, send an email to jupyter+unsubscr...@google

[jupyter] Re: build installable file - OLPC XO laptop - Fedora 18

Hello Tony, have a look at Miniconda. It has about 35 MB, instead of the 500+MB of a full Anaconda installer. Then create an Anaconda environment that has only the packages relevant for your environment: 1. Miniconda 2. Jupyter Notebook and dependencies 3. additional science packages Then creat

[jupyter] Re: Restart snippet numbering in the middle of a notebook?

On Tuesday, December 26, 2017 at 3:42:23 PM UTC+1, insearcho...@gmail.com wrote: > > But once I ship out the notebook to others, they'd need to do the same as > they execute cells, correct? > Once you ship out the notebook to others, they can execute and re-execute cells in whatever order they

[jupyter] Re: Working on a Common Lisp Jupyter notebook server - and experiencing a crash in ipython code

On Monday, December 18, 2017 at 3:00:23 PM UTC+1, Christian Schafmeister wrote: > > I fixed the problem last night - it turned out to be a threading issue. > When I rapidly evaluated notebook cells (hitting shift-enter really fast) > the Python in Jupyter Notebook would crash as shown. > Multiple

[jupyter] Re: Restart snippet numbering in the middle of a notebook?

Restart the kernel before each section. That will clear all state in memory, and reset the execution count. -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To unsubscribe from this group and stop receiving emails from it, send an email to j

[jupyter] Re: Working on a Common Lisp Jupyter notebook server - and experiencing a crash in ipython code

Hello Christian, the architecture looks like this: Browser <-> Jupyter Notebook <-> Kernel The browser communicates with the Jupyter Notebook server using HTTP and Websockets. The notebook server communicates with your kernel using ZMQ connections. In other words, the notebook server translates

Re: [jupyter] NotebookApp configuration error

If your editor shows 'c.NotebookApp.ip', but the error message is about 'NotebookApp.ip', then you're editing the wrong file (or you forgot to save the change, which has happened to me occasionally ;-). Maybe you installed Jupyter in multiple environments, or you copied the configuration file to

[jupyter] Re: How to resolve the port conflict for Jupyter Notebook and Hue? Thanks.

1. Jupyter is listening on localhost, but not on the external IP address of "myhostname". 2. If your cluster has a firewall, make sure that it exposes port 8880. hope that helps, Roland -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To u

[jupyter] Re: Jupyter notebook

Hello twanmal, could you please explain the problem you encountered? If the Anaconda environment is correctly initialized, the notebook server in that environment should have no problem to start with a localhost setup. With correctly initialized, I mean that you executed "source activate " (or

[jupyter] Re: Is it possible to run multiple Jupyter versions on a hub?

Kenneth, your notebook app in the alternate environment might be looking for its configuration in a different location. Make sure that it finds something valid :-) -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To unsubscribe from this gro

[jupyter] Re: Checkpoint curiosity

Hello David, on Data Science Experience, we call them "versions". When somebody shares a link to or schedules execution of a notebook, it's the last version/checkpoint that will be rendered or executed by default. Users can continue to edit and save their notebooks, that doesn't change what's

[jupyter] Re: Jupyter architecture and kernel development and extensibility...

> > >1. what is the role of kernels in Jupyter (how is the interaction with >other components) > > The general concept is... Browser/Client ---\ /---Kernel Browser/Client *Jupyter*Kernel Browser/Client ---/ \---Kernel where "Jupyter" can be a notebook server or k

[jupyter] Re: Written permission from Jupyter Notebook Copyright Holder

Hello, Disclaimer: Without being a lawyer, and not giving binding legal advice... We understand that it mentions that we will need to have a written permission > from Jupyter Notebook Copyright Holder > > before we endorse or promote derived product. > Your understanding is incorrect. Th

[jupyter] Re: Kernel error when connecting to a bash kernel in notebook

I found several references to this TypeError, all of them related somehow to unicode strings. Check your environment variables for unicode characters. If you find some, unset those variables before starting Jupyter. You can also try to run Jupyter with Python 3, which handles strings differently

[jupyter] Re: To what degree can users modify the notebook environment from within a Jupyter notebook?

They can access the address space of the Jupyter process through the /proc pseudo file system, can't they? I cannot tell how much effort it would be to mess with the process without crashing it. Given enough motivation, I'm sure it can be done. -- You received this message because you are subs

[jupyter] Re: How do I connect to the kernel using lldb -p without jupyter immediately restarting the kernel

When debugging, you could change the KernelRestarter class. I'm not sure if it's configurable, but you certainly can hack the code. It's in the jupyter_client package. hope that helps, Roland -- You received this message because you are subscribed to the Google Groups "Project Jupyter" grou

[jupyter] Re: To what degree can users modify the notebook environment from within a Jupyter notebook?

Hello Hussein, in a typical Jupyter installation, users have write access to the configuration files and kernel specs. They can wreak all kinds of havoc there. Even if you stop them from messing with configuration files and kernel specs, they may be able to install at ~/.local/ different versio

[jupyter] Re: recognizing updated kernels

Hello Gideon, if your kernel spec is hardwired to the old version, you'll have to replace the kernel spec with a new one. If your kernel spec just generally points at a "Julia environment" and uses whatever version happens to be installed there, it should work without changes. Unless it has a v

[jupyter] Re: plot draw

Have you tried setting the markersize? http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.plot Is the missing library problem? > What missing library? You didn't show the code you used for plotting, nor any error messages indicating a missing library. If it wasn't for a bit of text in

[jupyter] Re: Single kernel handling multiple projects

Hello Didier, the recently merged kernel discovery mechanism might be what you are looking for: https://github.com/jupyter/jupyter_client/pull/261 The idea is to implement kernel providers that can detect for example conda environments, Python virtual environments, or in your case Clojure proj

[jupyter] Re: Help

Hello Nicholas, it's much easier to spot a problem when there's code to look at :-) X = input("please input integer to add or Q to quit") > Do you even see the prompt? I'm not sure whether console-like input is supposed to be working in the web UI. cheers, Roland -- You received th

[jupyter] Re: Help

Hello Nicholas, On Sunday, October 8, 2017 at 9:10:00 PM UTC+2, Nicholas Rondon wrote: > > I was writing code in python and when I ran the code, it wouldn't run and > it showd an asterisk where it shows the number on the cell. Any help please? > Sounds like you code is entering an endless loop.

[jupyter] Re: Execute code on a remote kernel using HTTP with kernel_gateway

Hello Andrew, since you want to send the code for execution, you will have to run the kernel gateway in websocket mode. That means it's not a simple HTTP/REST call with "execute this" in the request and "here's the result" in the response. The sequence is more like: 1. Check if an instance of

Re: [jupyter] Re: Trouble accessing local notebooks on firefox when disconnected from internet

On Monday, September 25, 2017 at 6:23:13 PM UTC+2, takowl wrote: > > No, I don't think we're relying on any JS libraries fetched from the > internet. > My mistake. It's nbconvert, not notebook, that refers to MathJAX by CDN. -- You received this message because you are subscribed to the Google

[jupyter] Re: Trouble accessing local notebooks on firefox when disconnected from internet

Hello Florian, the browser probably pulls some JavaScript libraries from online locations. It seems that Chrome caches them more aggressively than Firefox. To verify, you can go offline and clear the browser cache in Chrome. It should then stop working, too ;-) cheers, Roland -- You receiv

Re: [jupyter] jupyter lab on docker to connect to host (matlab) kernel

Hello Karsten, one problem you will encounter when using remote kernels is that the kernel doesn't have access to the file system you're seeing in Jupyter Lab. That's almost sure to confuse users. They can edit files in their browser, but then the kernel doesn't see them. The notebook files wil

[jupyter] Re: Docker Jupiter persistent user and password

Hello Piero, the clean solution is to set up an LDAP server and have the Docker container authenticate users with that. A hacky solution could copy/append user definitions from an external volume when the container starts up. hope that helps, Roland -- You received this message because you

[jupyter] Re: How to get the matplotlib inline output via jupyter_client

Hi Jeff, stdout will give you only text printed to standard output. Cell execution also yields an execution result, which can come in different mime-types. Have a look at that. For example, execute the following in a Python notebook to see the difference: print ('standard output') 'ex

[jupyter] Re: JKG - using prespawned kernels

Hi Kevin, until the notebook is known, we cannot tell what type of kernel will be needed. The notebook-http mode has a notebook in the configuration, so it knows which type of kernel to start. To achieve what you are thinking of, you'd need separate pools of prespawned kernels for every kernel

[jupyter] Re: uninstall jupyter

On Wednesday, February 24, 2016 at 2:07:24 PM UTC+1, Vishal Mehra wrote: > > 2. how to change working directory from syswow 64 > You mean the working directory of Jupyter? Change it before starting Jupyter. -- You received this message because you are subscribed to the Google Groups "Project J

[jupyter] Re: Issue running jobs in Jupyter

On Thursday, July 6, 2017 at 1:45:10 AM UTC+2, iamhasib wrote: > > We have pretty large servers (3 servers, all m4.2xlarge servers) and test > data volume is pretty tiny. > Have you checked the configuration of the containers that are started? If a container is limited to x MB of RAM, it doesn'

[jupyter] Re: Mido / Python / Jupyter

Please consult the Anaconda documentation on how to add libraries: https://conda.io/docs/using/pkgs.html >From a Python notebook, you can use the ! prefix to execute shell commands. I hope that works on Windows, too. If not, use a separate terminal to change your Anaconda environment, then resta

[jupyter] Re: New version of Jupyter breaks code to get notebook file path

I'm glad you could sort out the problem. On Monday, June 19, 2017 at 8:17:12 PM UTC+2, DG wrote: > > BTW, I could not find those "release notes" you mention, documenting > /api/sessions. Care to point me to the correct location? > It's not about /api/sessions in particular. Token authentication

[jupyter] Re: New version of Jupyter breaks code to get notebook file path

I think there were changes regarding authentication. You might have to pass a security token as a query parameter or header now. Try some command-line debugging with curl on the /api/sessions URL. If you know the old and new versions of the various Jupyter components, have a look at the release

Re: [jupyter] command line ipython kernel incantation / API?

Hello John, if you get a message ID instead of an actual reply, then you didn't specify to wait for the reply: https://github.com/jupyter/jupyter_client/blob/master/jupyter_client/blocking/client.py#L44 I'm not familiar with the API, but this might work: replymsg = c.execute('a=9', reply=True)

[jupyter] Re: Can the notebook server shut itself down?

Hi Brendan, regarding your fallback solution: the notebook server is Python code. You can import the packages and call the functions to start a notebook server in the current process. The 'jupyter' command itself is a Python script, so just follow the code path to see what it does when invoked

[jupyter] Re: how to enforce one running kernel per user?

Hello Igor, there's no config parameter for this, afaik. The kernel gateway has a parameter to limit the number of kernels, but the notebook server does not. And even the kernel gateway will not automatically stop a running kernel. Rather, it would refuse to create new kernels until an old one

[jupyter] Re: How to share a Jupyter installation for team notebook development

Hello Heath, if you want to centralize the library management for the kernel runtime, you could run the Jupyter kernel gateway on a shared server. The users could then run their own copy of nb2kg as a browser-based frontend to that gateway. However, I have no idea how you would manage file uplo

[jupyter] Re: how to import functions defined in other notebooks

Hello Andy, relative imports doesn't mean you can navigate to parent directories. Two suggestions: 1. Create a symbolic link (ln -s), so the other directory becomes a subdirectory. 2. Move the shared code to a library that can be used from all notebooks. hope that helps, Roland -- You receiv

[jupyter] Re: Any plan in how to limit the notebook instances resource usage of one Jupyter notebook server?

Realistically, you have to run either the whole Jupyter or each notebook kernel in a container (Docker, LXC,...). Then you can set limits on the container(s), and/or restrict the number of simultaneously running kernels through the Jupyter kernel manager. Without containers, a user can consume

  1   2   >