Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    Kal Sze
    @ksze
    Oh god, will Jupyter stop polluting my ~/.local/share? What is this? Writing notebook-signing key to /home/ksze/.local/share/jupyter/notebook_secret
    How do I control where it writes the notebook_secret?
    I mean, there is no guarantee that I will have a single JupyterLab installation or that I want to use the same notebook signing key for all notebooks and all JupyterLab installations.
    And then what is this? Notebook <redacted>.ipynb is not trusted
    I can't find any relevant info using the keyword "trust" in the JupyterLab documentation: https://jupyterlab.readthedocs.io/en/stable/search.html?q=trust&check_keywords=yes&area=default#
    6 replies
    Enquandriant
    @Enquandriant
    Would anyone here be able to help me with a TLJH problem? I am following the guide for getting it working on my own hardware, on a clean install of ubuntu focal, but when I go to the url and try and login as the admin account I just created I get redirected back to the login page. (And the jupyterhub logs indicate I logged in successfully)
    Renan
    @renanmb
    Hello!
    I am trying to run Codex on Jupyterlab on a GCP server but i get the following:
    [I 2021-09-22 19:00:03.921 ServerApp] 302 GET / (::1) 0.61ms
    [W 2021-09-22 19:00:07.030 LabApp] Could not determine jupyterlab build status without nodejs
    [W 2021-09-22 19:00:08.247 ServerApp] Notebook Untitled2.ipynb is not trusted
    [I 2021-09-22 19:00:08.739 ServerApp] Kernel started: c9114e1c-70b4-4791-b32d-35aae6dbc686
    [W 2021-09-22 19:00:12.470 ServerApp] 404 POST /jupyterlab-codex/completion?1632337212823 (::1) 26.81ms referer=http://localhost:8888/lab/tree/Untitled2.ipynb
    [W 2021-09-22 19:00:12.470 ServerApp] 404 POST /jupyterlab-codex/completion?1632337212823 (::1) 26.81ms referer=http://localhost:8888/lab/tree/Untitled2.ipynb
    For some reason it dont let me talk to OpenAI server
    unwaz
    @unwaz
    Hi, I've a noob problem: How do I access jupyterlab's active kernelspec.language and store it in a variable? I'd like to customise jupyterlab dependent on the kernel selected.
    Michał Krassowski
    @krassowski
    @unwaz for JupyterLab questions see https://gitter.im/jupyterlab/jupyterlab
    Vladimir Korolev
    @balodja
    Hi all, just a simple question about jupyter architecture. I'm looking at the server API and I don't understand how does the client interact with the kernel (executing the cells, asking for a autocomplete list and so on). It's not even clear from the network communication between browser and web server. Could someone point me to the right direction, please?
    9 replies
    Link to the right place in sources of the nbclassic or jupyterlab would be quite enough, I hope.
    Lee Martie
    @133martie
    We have deployed z2jh on IBM's Kubernetes services (15 large nodes), with about 20 active servers at any moment.. but noticed, in general, spawning a JupterLab server takes 30sec and up to 60sec. Is this typical and are there ways to improve the performance of the spawner? Thanks for any help!
    5 replies
    Tingkai Liu
    @TK-21st
    Not sure where to post this but can someone take a look at this Calysto/matlab_kernel#157 for matlab_kernel? We’re using matlab_kernel for a course and it’d be really helpful if we can use svg images since they would significantly reduce the notebook file sizes. Thanks!
    vincenzoml
    @vincenzoml
    Hi there, I've spent a full day configuring jupyterhub in a docker container, and I am up to a point where I have it properly configured with letsencrypt and one pam user. However, "spawning server" fails complaining about pam
    Can anyone help me? I understand maybe I should use something called DockerSpawner but hell if I understand where I can set such a thing up
    in which file I mean
    here is the log https://pastebin.com/TudrxC1N even if, honestly, I can't tell anymore it's complaining about pam; that's probably my mistake, pam seems no longer involved
    vincenzoml
    @vincenzoml
    I beg you please :) I've been wasting hours on this setup and have something like 30 years of expertise, of which 20 on linux, but still, it's hours I'm searching and I have no clue of what I have to do next
    Min RK
    @minrk
    @vincenzoml ModuleNotFoundError: No module named 'notebook' means the notebook package is not found. Try python3 -m pip install notebook.
    (possibly with sudo since it looks like you are using a system-wide install)
    vincenzoml
    @vincenzoml
    @minrk wow, that was it... I did not note that error.
    thanks
    Ari
    @ahribellah
    is there a way to disable Ctrl-X functionality in JupyterLab? trying to type APL characters (the IME keyboard uses Ctrl+<key> commands) and Ctrl-X keeps cutting the entire line I'm on. I can't seem to find the command in Keyboard Shortcuts settings, though. all I see is X to cut cell
    Ari
    @ahribellah
    Okay, I think I figured out how I'm supposed to be able to disable it - fileeditor:cut and .jp-FileEditor, but I've put that into user preferences, added "disabled": true for it and several other key combos and nothing is happening. I've tried Ctrl X, Accel X, and even "" under "keys", but I cannot seem to disable it
    Joseph D. Long
    @joseph-long
    Hi all, I'm following https://github.com/jupyterlab/jupyterlab-mp4/blob/master/tutorial.md in hopes of making an extension that can explore a large file using server-side code and display a summary. I think I still want to hook in to the mime render infrastructure, since I want it to work when you double-click a file with a certain extension
    1 reply
    The question I have is how to avoid moving the whole file across when using the IRenderMime.IExtension interface... there's a dataType property that can be "string" or "json" but I'm pretty sure what I want is... none?
    And then I can move some summary info across using a request to an endpoint on the backend
    demir kulo
    @blessedbck_twitter
    hello! im an newbie. my question: whats the easiest way to import an colab notebook and make it work-because i wanted to rent a gpu and the support told me it works on jupyter lab.
    1 reply
    Sandhya Govindaraju
    @sandhya-sago
    Hi! I am developing a Jupyterlab Extension using React and Typescript. Couple of questions wrt styling:
    1. Does Jupyterlab uses some standard theming so that I can reuse them - like Material UI/ Bootstrap UI, etc?
    2. Is there some resource/tutorial for using jupyterlab ui components? For example, I am trying to use caretRightIcon but do not know how it looks like. It will be nice to have a place to see it icons.
      Thanks!
    2 replies
    Iwan Aucamp
    @aucampia
    Hi, what is the best practice to make models that are reproduceable from Jupyter? Would it be to commit the notebook and run CI/CD on it?
    3 replies
    Iwan Aucamp
    @aucampia
    Maybe I should start from a broader problem. We want to allow data scientists to deploy ML models to production for online experimentation. But for this i feel the minimum requirement should be that it is reproducible, as otherwise I don't see how it is helpful to test something you can't identify. And if it works well and you want to change it, then that is not exactly an option if you can't reproduce it.
    7 replies
    Iwan Aucamp
    @aucampia
    To explain a bit further, If the data scientists are making these ml models with jupyter however and just directly exporting them from jupyter as they are working on it then I don't know how we can be sure about the state and inputs. Even if I were to save the notebook at this point in time that won't store all inputs and state. So the only real way I see to use jupyter to make reproducible ml models would be to commit it to git and run some CI pipeline which provides all inputs. Of course it could still not be reproducible but this is at least one step in that direction.
    Iwan Aucamp
    @aucampia
    thanks for the inputs @bollwyvl - will think about it a bit
    3 replies
    Renan
    @renanmb
    Other than ROS, are there any simulation environments for robotics that have great documentation and work well on jupyter ?
    Zachary Sailer
    @Zsailer
    The future of the classic notebook interface—jupyter/notebook#6210
    danx0r
    @danx0r
    hi -- I need a way to populate a new user's account with some local data
    I'm looking into building a plugin unless something similar already exists
    Chandra Vedantham
    @Chandu
    Team, I am trying to build an extension for jupyterlab (using https://github.com/jupyterlab/extension-cookiecutter-ts) and came across jlpm command. Should I definitely use jlpm? Can we stick to npm instead of jlpm. I could not find much documentation about jlpm and why would one use jlpm over npm.
    Michał Krassowski
    @krassowski
    It's an alias for yarn; there were some discussions about dropping it, but realistically it stays for now and I would encourage using it (there is lots of documentation for yarn out there); there is a dedicated channel for lab btw: https://gitter.im/jupyterlab/jupyterlab
    Chandra Vedantham
    @Chandu
    Thanks @krassowski .
    danx0r
    @danx0r
    OK, I take it no one has heard of such a tljh plugin (to populate user home directories with data).
    I will therefore embark on attempting to build my own
    Michał Krassowski
    @krassowski
    @danx0r you may also want to check on discourse (https://discourse.jupyter.org/), but I would add more details as I do not get what you are trying to do from this very brief description (I mean I imagine a few possibilities).
    danx0r
    @danx0r
    hi @krassowski thanks for replying. What I want is pretty simple: when signing up a new user, I want to put some files and folders in their home directory
    the pluggy hook tljh_new_user_create looks like what I need
    danx0r
    @danx0r
    a tljh pluggy plugin to load data & run a script at onboarding of new users
    sabs99
    @sabs99:matrix.org
    [m]
    i have a ConnectionRefusedError and don't know how to go about it (anaconda3, total beginner^)