Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    jattack03
    @jattack03
    How would i perform a calcluation with 2 values on the same row of a csv then do the same thing for the next row and so on
    L. Kärkkäinen
    @Tronic
    @jattack03 Use df = pandas.read_csv("somefile.csv") to read your data, then get the columns you want out of df, and do the calculation, something like df["Column A"] * df["Column B"]
    Dr. Muhammad Anjum
    @anjumuaf123_twitter
    @tanimislam Hi Sir, I need your help for python learning
    Dr Paul Brewer
    @DrPaulBrewer
    @parente This SO question demonstrates a numpy floating point inconsistency where order of operations may be affecting the math. It seems to affect jupyter more than ipython 5.5.0. I'm mostly curious and put a bounty on it, the OP has the actual problem.
    Wolf Vollprecht
    @wolfv
    @zachetienne with jupyterlab you have that, and you can also open a code editor with the classical notebook i believe
    Zachariah Etienne
    @zachetienne
    @wolfv : That may be true, but jupyterlab breaks my notebooks (which use raw HTML to make clickable Tables of Contents). Also I'm not familiar with jupyterlab having C syntax highlighting capabilities in code cells when a Python kernel is active (I want to run Python code in code blocks but edit & save C code in these blocks as well)
    Wolf Vollprecht
    @wolfv
    ok, now I see a little clearer what you want. This is probably also not supported in JupyterLab except maybe for markdown
    where you might get the right highlighting by adding ```c
    in front of the code block
    Zachariah Etienne
    @zachetienne
    @wolfv : Yep, the markdown syntax highlighting is great. My project (https://github.com/zachetienne/nrpytutorial) requires that both Python & C code be developed side-by-side, but only Python need be executed by the running Jupyter kernel.
    Zachariah Etienne
    @zachetienne
    ... hence the desire for C syntax highlighting in a code block, e.g., when the %%writefile blah.c magic is used
    Krzysztof Nazarewski
    @nazarewk
    is there some package/plugin for easily displaying interactive tables? I am not using pandas or anything like it, just plain dicts and lists
    Min RK
    @minrk
    I'm not sure how easy it is to automatically infer the structure of a table from a dict or list, so I'm not sure a publicly distributed one exists. You can write your own custom formatter for your table-style, though.
    Perhaps the simplest version, though, is to rely on pandas' own display implementation and use pandas.from_dict just for display.
    Krzysztof Nazarewski
    @nazarewk
    i guess i will just install pandas as everything is written for it ;/
    Min RK
    @minrk
    In that notebook, I wrote a quick and dirty example of a display formatter for dict representations of tables that you can use, but installing pandas is probably often the easiest route.
    Jared Vacanti
    @jaredvacanti
    I'm having some trouble with the most recent image of jupyterhub/configurable-http-proxy - i'm not sure but I think the new image exposing multiple ports may have something to do with it. Before I believe only 8000 was exposed, now 8001 as well. Where should I direct traffic?
    Min RK
    @minrk
    What sort of issue? I wouldn't expect exposing ports to do that.
    8000 is the incoming traffic port (unchanged), and 8001 is the API port (also unchanged)
    Jared Vacanti
    @jaredvacanti
    I think it has to do with my service discovery, going to try to dig into it more. So external traffic should direct to 8000 ?
    because it does work fine when using 4.1.0
    Min RK
    @minrk
    Yeah, I wouldn't expect any changes regarding ports, so please do open an issue if you find something changed that maybe shouldn't have.
    Maybe we shouldn't really be exposing port 8001
    EXPOSE directives in images are a little silly imo, I'm not sure what problem they really solve
    Jared Vacanti
    @jaredvacanti
    I use EXPOSE directives with registrator to pickup services. EXPOSE isnt necessary for communication between containers otherwise
    ill report if I find out what's the issue, but I'm sure it's along those lines
    Min RK
    @minrk
    so if we only EXPOSEd the public-traffic port, your discovery would work?
    Do you use discovery to find the api port as well?
    I'd be interested to see how you use it if any of it is public, I've never really understood when EXPOSE belongs in the Dockerfile vs the service description at runtime.
    Jared Vacanti
    @jaredvacanti
    yes, I think it would work and it works fine now when I use the older docker image version. I think it has more to do with how it is proxying it through, because i'm getting a not found. I have also recently made some nginx adjustments. I've got some logs to dig through
    Min RK
    @minrk
    not found is exactly what you'd get if you hit the api port instead of the public port, fwiw
    Brenton Bills
    @fud
    Hi, wondeing if jupyterlab changed to jupyter-lab as the command and trying to follow a tutorial that has kernelspec as an argument, but I just don't see that argument. I am using python 3.8 using pyenv and installed 1.2.3 jupyterlab with pipx
    Vladimir Lipkin
    @vhaldemar
    Hello!
    I want to customize the page to redirect after JupyterHub spawn and I want to customize it per-user via Spawner.
    There are Spawner.default_url attribute, but as far I can see, it doesn't do anything (except of cmd/args modification).
    Is there are way to customize the thing that I want?
    Thank you!
    Christopher Ostrouchov
    @costrouc
    Hi all. I was wondering if it is possible to restart a notebook from within python and not the "Select Kernel" interface? Ideally I would like to do restart_kernel(name='foobar')
    Sean McNamara
    @cmsolutions_gitlab
    Hi all. I'm trying to install some extensions (plotly) into jupyter lab, and the webpack build (jupyter lab build) has been running at 100% cpu for a couple of hours. I found a few posts online talking about hanging builds but didn't find a resolution. It's not failing, it's just not finishing... Any pointers?
    Joe Quigley
    @quigleyj-mavenomics
    Does --minimize=False help?
    Sean McNamara
    @cmsolutions_gitlab
    @quigleyj-mavenomics I'll have to check. I reverted to using notebook vs lab and everything works just fine. I left the lab build running for almost 20 hours and it still hadn't completed.
    Joe Quigley
    @quigleyj-mavenomics
    At 20 hours, I almost wonder if it was just spinning waiting for a lock on something... If you can reproduce it, might be worth filing an issue on the JupyterLab github- builds shouldn't take more than a few minutes unless they're thrashing the swapfile, and even then, builds don't stretch out more than an hour.
    Nazmican Çalık
    @nazmicancalik
    Hi, I am experiencing a problem and thought here it might help me find a solution. I have anaconda 3.7 installed on my ubuntu 18.04. I start the jupyter notebook for a deep learning task. I load an image dataset (cifar-10) 303 mb. When I first run my code there is no problem. I train the model, however if I run it again (after restart kernel and run all) some cells which took miliseconds to run in the first run crashes the system and I don't get a response. Is there anybody who had a similar situation or has an idea. PS: I don't have any extensions (In some cases Variable inspector extension was the problem) Thanks
    Nazmican Çalık
    @nazmicancalik
    Hi to everybody: I have found the solution - one of the packages in the system was causing the memory overload. When I updated my system everyth
    everything worked*
    emmanuellbadea
    @emmanuellbadea
    hello
    kdeepak99
    @kdeepak99
    @kdeepak99
    @kdeepak99
    Guys I am getting below error
    Traceback (most recent call last):
    File "/usr/local/Cellar/jupyterlab/1.2.0/libexec/bin/jupyter", line 6, in <module>
    from pkg_resources import load_entry_point
    ModuleNotFoundError: No module named 'pkg_resources'
    while starting jupyter notebook
    plz help me how to resolve issue
    @benzrf_twitter
    hey—while installing jupyter on my professor's laptop, it gets to ipykernel, runs the tests for it, and then for the asyncio tests, it just hangs and then crashes after a while
    i don't have the error on me atm :|
    he uses macos, & i was installing it via nix
    anyone know what that might be?
    golf-player
    @golf-player
    Q: Is there such a thing as a service that runs nbconvert and caches results based on papermill parameters?
    I assume it would be difficult since jupyter is unaware of whether something is run with papermill or not