When I use iPython to run some scripts, I've got the following warning message in iTerm:
/Users/###/anaconda3/lib/python3.6/site-packages/ipykernel/displayhook.py:12:
VisibleDeprecationWarning: zmq.eventloop.minitornado is deprecated in pyzmq 14.0 and will be removed.
Install tornado itself to use zmq with the tornado IOLoop.
When I try to run my code in Visual Studio Code, I've got the following error message:
Error message in VSCode
I have already tried to update Anaconda, I have already tried to install nomkl with conda as well, without any success.
iTerm and VisualStudio on Mac OS 10.14.6, Anaconda up to date (4.7.12)
Remedy Option A)
Use anaconda-tools to separate an environment, say aLastWorkingENV, where you make use of anaconda-tooling to roll-back all components to a last working state, which does not whine about future deprecations. In such aLastWorkingENV environment, all thus synchronised programs will work smooth (as they did before a first component update, that has moved the stick)
Remedy Option B)
Use anaconda-tools to update all current components and all their supporting modules and all packages versions [ python, ipython, pyzmq, jupyter, ..., ] to their most recent versions, where products QA coverage ought either confirm the PASS, or provide a remedy or workaround or the last-trouble-free version ( to eventually downgrade one such package, based on such exception, if present and relevant )
Either way, your code would stop throwing the soft-warnings and remain in such state, until any un-coordinated package update takes place.
Related
I would like to use the package HDF5
In my Pluto.jl, I have the line
using HDF5
When I try to evaluate this cell, I get the error message
"ERROR: LoadError: HDF5 is not properly installed. Please run Pkg.build("HDF5") and restart Julia."
I would like to do this, but when I go to the terminal, I can't do this while I have Pluto open.
I've tried running Pluto in the background with a command like
Pluto.run() &
But this code is completely wrong.
I've also heard that there sometimes appears a cloud icon above the cell, which would allow me to download HDF5 directly.
In any case, it seems to me like any time this happens, I will have to write down which package I need to install, and then kill my Pluto notebook, go to Julia, install, and restart Julia. Surely there is a better way? Can anyone help me find it?
When the package is correctly installed or could be installed without problems, using HDF5 in Pluto itself is sufficient. The built-in Pluto package manager takes care about the installation.
There are edge cases where due to issues with external packages installation does not work out-of-the-box. In this case, it could help to install the package in a temp environment before starting Pluto:
] activate --temp
] add HDF5
follwed by whatever steps required to get the package working in Julia itself, like re-building it.
This should really be a workaround and should be fixed in the corresponding package - consider creating an Issue there if it does not exist already.
macOS Big Sur 11.3.1 elixir 1.9.4 erlang 22.0
I am brand new to Elixir and I am trying to run a debugger for the first time. I run my webserver's development environment with iex -S mix and get the interactive elixir prompt. Then I type :debugger.start() and receive an immense output and then erlang crashes.
This is a small snippet, but the actual output is huge.
== NULL" failed in Register(): Class "wxTextValidator" already in RTTI table - have you used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some object file twice)?
I have to use the versions listed at the top of the post, the project is version locked. Does anyone have any insight into what is causing this? It may be related to what I had to do to install erlang to begin with. But I'm not sure.
Thank you in advance.
I have never used the debugger, but I believe it relies on some of the same components as the observer. I would recommend installing Elixir and Erlang via asdf -- that makes it easier to install specific versions.
Once you've got things installed, you can use homebrew to install wxmac
brew install wxmac
Then to your mix.exs, add :observer, :wx to your extra_applications, e.g.
extra_applications: [:logger, :observer, :wx]
I have created my own Anaconda environment. I needed to use the 'comtypes' package to connect to the software ETABS, so I first downloaded it with the conda installation command and then I tried the next command line:
helper = comtypes.client.CreateObject('ETABSv17.Helper')
But, I received the next error:
AttributeError: module 'comtypes.gen.ETABSv17' has no attribute 'cHelper'
However, I realized that if a try the same command line with the base environment coming with Anaconda, that includes a pre-installed 'comtypes' package, there is no problem. In fact, if I check comtypes.gen.ETABSv17 with my customized environment and with the one from Conda, there is a huge difference between available attributes, even when they have the same version number: comtypes 1.1.7.
How can I get the exactly same 'comtypes' package that comes with Anaconda in order to have all the associated attributes working? Maybe, could there be some package dependence that I am missing?
Pd: I am using Windows10 and I have tried it in Pycharm and Spyder with Python 3.7
Thank you all !
I had this issue and i was using python 32 bit version. So uninstalled and installed a 64 bit version and now its working fine. May be this helps
How do you get LaTex output on Spyder? I saw some other post recommending installing the MikTex package but I don't know how to use it.
I installed the MikTex package on Anaconda Spyder, but now how do I use it? I have:
import sympy as sp
sp.init_printing()
x = sp.symbols('x')
integrate(x)
But nothing
I had troubles getting Latex output to work in Spyder too, and I was able to hunt the bug down. At least for me, the Miktex installation delivered by the conda miktex package got confused about being a shared or a single-user installation, and consequently failed to install the necessary packages on-the-fly.
Here's my strategy:
Verify if you have the same problem by looking at the Miktex log files (on Windows, in your user's AppData\Local\Miktex\2.9\miktex\log directory); watch out for lines like
installing package (...) triggered by (...), immediately followed by entries marked FATAL with the error message Administrative mode cannot be enabled (makes no sense) because this is not a shared MiKTeX setup.
If this is the same issue for you, start the Miktex console in the conda Miktex package; the location is (Anaconda dir)\Library\miktex\texmfs\install\miktex\bin\miktex-console.exe. (Trying to run the _admin version fails with a similar "Administrative mode" error message.)
In the console, go to "Packages", and manuall install the following packages: standalone, varwidth and zhmetrics
After these steps, Latex output worked for me. Note that these package installations are user-specific, so you'll have to repeat them for each user.
After making cabal install of the darcsden code I get this message:
cabal: The following packages are likely to be broken by the reinstalls:
bin-package-db-0.0.0.0
ghc-7.4.1
Use --force-reinstalls if you want to install anyway.
How do I get around this? What does it mean?
Why does it happen?
If you look at the full output of cabal install darcsden, you will find several lines that look like this:
binary-0.5.1.0 -bytestring-in-base (reinstall) changes: array-0.4.0.0 ->
0.3.0.3, containers-0.4.2.1 -> 0.4.1.0
This means that cabal has found an install plan that involves (destructively) reinstalling packages that you already have on your system.
Now, GHC packages are rather sensitive when it comes to their (reflexive) dependencies, and generally only work if exactly the right version of all dependencies is available, compiled against the right versions of their dependencies and so on. Therefore, replacing an already installed package with a new version of changed dependencies can cause some packages on your system to become unusable. Since version 0.14.0, cabal warns you about such a situation in advance to prevent you from accidentally breaking your system.
In your case, ghc and bin-package-db are among the potentially broken packages, because they depend on binary which gets reinstalled. So you should not try to use the --force-reinstalls flag, because it might really break your GHC.
What can you do?
If you scan what is going to be reinstalled, you see that quite a few dependencies are downgraded. This hints at the fact that the package you are trying to install might not be properly updated to GHC 7.4.1 yet.
You can in general try to call cabal install darcsden --avoid-reinstalls to explicitly try to find an install plan that has no reinstalls. Unfortunately, in this case, it fails (for me).
I've briefly looked at the darcsden package description, but it looks like quite a few dependencies of darcsden need to be updated. So the remaining options are: Convince the author(s) of darcsden to release an updated version, or install darcsden using an older version of GHC (such as 7.0.4), which should just work.