How do I change the kernel/python version for iPython? - python

I have installed iPython using pip in OS X 10.10, and it gave me the "ipython" and "ipython2" commands, which run great, but which use OS X's default python version 2.7.9. I downloaded and installed the latest release of Python3.4 and can load it with the command "python3," but cannot find a way to get iPython to use this version of python. The iPython Web site states the package can be used with python versions 3.3 and above, but I cannot find any instruction on how to change the default python version used.
So far I have found that the jupyter package for iPython has a kernel specification in /usr/local/share/jupyter/kernels/, which is just a folder called "python2" containing a json file that points to the system's python 2.7.6, but altering this to point to the new python3.4 installation does not work. My guess is this configuration is for the ipython notebook.
I've also tried the approach here: ipython reads wrong python version
In doing so I've duplicated the ipython2 command in /user/local/bin/ and edited it to use the python3 interpreter located at /Library/Frameworks/Python.framework/Versions/3.4/bin/python3, however, this gives me an error "ImportError: No module named 'IPython'," which suggests the python3 installation does not have ipython installed.
To tackle this, I've tried uninstalling ipython and reinstalling it using pip, but it just targets the system's Python 2.7 installation and does nothing for python3.
Does anyone know how to configure iPython to use a different python version, or even install a separate ipython installation for python3? Ultimately it would be nice to quickly switch back and forth depending on my needs.

I just found the answer. In essence, this stems from not understanding the python installation layout and how resources are separated between installed interpreters. It appears each python version will have its own repository of tools, and the current "pip" command I had installed on the system was mapped for use with python 2.7, so all libraries, tools, and other details it managed where available only to python 2.7. This included iPython.
I thought that installing python3 would make all these libraries available to the new interpreter by default, but it appears the system keeps them all separate (which makes sense). The main issue here was continuing to use "pip" which targeted the old installation, instead of the new "pip3" command included with python3. By using pip3 to reinstall iPython, I was able to get iPython3 installed properly.
Unfortunately this setup means needing to re-download other libraries, but that's not too difficult.
It might seem like a trivial issue in hindsight, but this had me completely stuck. I hope this helps someone else in this situation.

Related

Installing pip with default python on Mac OS 10.14 Mojave

I would like to install pip for the default installation of Python on Mac OS.
Please don't recommend brew, I already have it and installed Python 3 with it, but it seems that Automator only knows how to use the default version of Python located in /usr/bin/python That's the reason behind my specific request
I did my homework first, or tried to, before asking the question, but what I found confusing is that the recommended method seems to be using get-pip.py, but the pip documentation says
Warning Be cautious if you are using a Python install that is managed
by your operating system or another package manager. get-pip.py does
not coordinate with those tools, and may leave your system in an
inconsistent state.
This threw me off, as I don't want to risk breaking the default Python on Mac OS, as I understood that might mess my system.
I also didn't want to use the deprecated easy_install.
And I couldn't find an answer to my question, as usually the answers just recommend installing a different version of Python with brew.
Please don't recommend brew, I already have it and installed Python 3 with it, but it seems that Automator only knows how to use the default version of Python located in /usr/bin/python That's the reason behind my specific request
Can you possibly use "Run Shell Script" in Automator and specify the python version you want to use. See Specify which version of Python runs in Automator? and https://apple.stackexchange.com/questions/233890/calling-python-3-script-from-applescript
Problem
Seems like Automator isn’t loading /usr/local/Cellar/bin into your PATH. You can echo $PATH in Automator to confirm this.
Solution
Reinstall using brew and ensure that you run brew link python.
You can export PATH=... before running your script or move /usr/bin/python to /usr/bin/pythonx.x where x is the default version installed, then symlink /usr/bin/python to your brew installed python in /usr/local/bin/.

IDLE and python is different, not able to install modules properly

thanks for reading this. I am using macOS High Sierra. I am not very familiar with terminal or environment variables, but am trying to learn more. From reading other threads and google, it seems like I either have multiple pythons installed, or have pythons running from different paths. However I am not able to find a solution to resolving this, either by re-pathing my IDLE or deleting it entirely.
I do have python, python launcher, and anaconda (not very sure how anaconda works, have it installed a few years back and didn't touch it) installed. I am trying to install pandas (pip install pandas), which tells me that I have it installed, but when I run it on IDLE, it says module not found. Though if i run python3 on terminal and type my code in, it works (so pandas has indeed been installed).
When i run which python on terminal, it returns
/Users/myname/anaconda3/bin/python
(when i enter into this directory from terminal, it shows that in the bin folder, I have python, python.app, python3, python3-config, python3.7, python3.7-config, python3.7m, python3.7m-config)
When i run which idle on terminal, it returns
/usr/bin/idle (im not even sure how to find this directory from the terminal)
When i run import os; print(os.path) on IDLE, it returns module 'posixpath' from '/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/posixpath.py'
Would really appreciate some help to figure out how to ensure that when i install modules from terminal, it would be installed into the same python as the one IDLE is using. Also, I would like to know whether it is possible for me to work on VSCode instead of IDLE. I cant seem to find suitable extensions for data science and its related modules (like statsmodels, pandas etc). Thanks a lot!
First of all, a quick description of anaconda:
Anaconda is meant to help you manage multiple python "environments", each one potentially having its own python version and installed packages (with their own respective versions). This is really useful in cases where you would like multiple python versions for different tasks or when there is some conflict in versions of packages, required by other ones. By default, anaconda creates a "base" environment with a specific python version, IDLE and pip. Also, anaconda provides an improved way (with respect to pip) of installing and managing packages via the command conda install <package-name>.
For the rest, I will be using the word "vanilla" to refer to the python/installation that you manually set up, independent of anaconda.
Explanation of the problem:
Now, the problem arises since you also installed python independently. The details of the problem depend on how exactly you set up both python and anaconda, so I cannot tell you exactly what went wrong. Also, I am not an OSX user, so I have no idea how python is installed and what it downloads/sets alongside.
By your description however, it seems that the "vanilla" python installation did not overwrite neither your anaconda python nor anaconda's pip, but it did install IDLE and set it up to use this new python.
So right now, when you are downloading something via pip, only the python from anaconda is able to see that and not IDLE's python.
Possible solutions:
1. Quick fix:
Just run IDLE via /Users/myname/anaconda3/bin/idle3 every time. This one uses anaconda's python and should be able to see all packages installed via conda install of pip install (*). I get this is tiresome, but you don't have to delete anything. You can also set an "alias" in your ~/.bashrc file to make the command idle specifically linking you there. Let me know with a comment if you would like me to explain how to do that, as this answer will get too long and redundant.
2. Remove conda altogether (not recommended)
You can search google on how to uninstall anaconda along with everything that it has installed. What I do not know at this point is whether your "vanilla" python will become the default, whether you will need to also manually install pip again and whether there is the need to reinstall python in order for everything to work properly.
3. Remove your python "vanilla" installation and only use anaconda
Again, I do not know how python installation works in OSX, but it should be reasonably straightforward to uninstall it. The problem now is that probably you will not have a launcher for IDLE (since I am guessing anaconda doesn't provide one on OSX) but you will be able to use it via the terminal as described in 1..
4. Last resort:
If everything fails, simply uninstall both your vanilla python (which I presume will also uninstall IDLE) and anaconda which will uninstall its own python, pip and idle versions. The relevant documentation should not be difficult to follow. Then, reinstall whichever you want anew.
Finally:
When you solve your problems, any IDE you choose, being VScode (I haven't use that either), pycharm or something else, will probably be able to integrate with your installed python. There is no need to install a new python "bundle" with every IDE.
(*): Since you said that after typing pip install pandas your anaconda's python can import pandas while IDLE cannot, I am implying in my answer that pip is also the one that comes with anaconda. You can make sure this is the case by typing which pip which should point to an anaconda directory, probably /Users/myname/anaconda3/bin/pip
First: This would be a comment if I had enough reputation.
Second: I would just delete python. Everything. And reinstall it.
To repeat and summarized what has been said on various other question answers:
1a. 3rd party packages are installed for a particular python(3).exe binary.
1b. To install multiple packages to multiple binaries, see the option from python -m pip -h.
To find out which python binary is running, execute import sys; print(sys.executable).
3a. For 3rd party package xyz usually installed in some_python/Lib/site-packages, IDLE itself has nothing to do with whether import xyz works. It only matters whether xyz is installed for 'somepython' (see 1a).
3b. To run IDLE with 'somepython', run somepython -m idlelib in a terminal or console.
somepython can be a name recognized by the OS or a path to a python executable.

'python' points to anaconda installation

I've installed anaconda, but I still want to use python as normal within a Terminal window (zsh). At the moment, running which python shows /Users/USERNAME/opt/anaconda3/bin/python.
I don't want this command to point to my anaconda installation. I want to use python as I have in the past, and for it to point to my python 2.7 installation (/usr/bin/python).
The same issue is present with pip and with python3. How do I change these paths back to their default locations?
I want to start out by saying I personally have never used Anaconda, but I have many friends who have and they always seem to wind up getting into hot water as far as configurations go. If you don't need the data-science specific tools that Anaconda offers (most of which I imagine you can get just as easily by using pip install <package_name>), then I STRONGLY urge you to not use Anaconda. I would suggest you instead use homebrew to install python 3.7 and leave the system version of python 2.7 alone... Anyways...
I recently fixed an issue similar to the one you're describing for a friend. I ended up deleted his anacaonda build and reinstalling fresh from homebrew (link: https://brew.sh/), so this is how I helped him and hopefully this will help you too.
Firstly, I would check the configuration of your ~/.bash_profile (or potentially ~/.bashrc) file. Comment out any lines you find that have to do with Anaconda by preceding with the # character. I believe when anaconda installs it puts the location of the anaconda-version-of-python higher in the PATH hierarchy, therefore you computer sees it first before it has the chance to look in /usr/bin...
Secondly, you need to move the anaconda-python to the trash... I would start by trashing the whole '/Users/USERNAME/opt/anaconda3' directory. If memory serves me correctly anaconda stores other python related things in the /MacitoshHD/Library/Frameworks directory, but don't quote me on that. I would advise looking there and trashing whatever seems to be related to the anaconda install.
After that, in a new terminal (so that your PATH refreshes), I installed homebrew, and then ran the command brew install python3 to install python3. I think perhaps I ran the command recommended in the installer output, something like "brew link python3" to symlink python3 where it needs to be linked after the installation.
Finally, everything is done python2 and python3 are installed and working with no issues. python2 is the system default version and python3 is the homebrew installed version. Feel free to ask any questions and I will try to help more, hopefully those instructions will get you started though.

How do I get python to import pandas?

I installed Python 3.5.1 from www.python.org. Everything works great. Except that you can't install pandas using pip (it needs visualstudio to compile, which I don't have). So I installed Anaconda (www.continuum.io/downloads). Now I can see pandas as part of the list of installed modules, but when I run python programs I still get:
ImportError: No module named 'pandas'
How do I set up my environment to use the modules from Anaconda?
Note: I have Anaconda's home directory and Library/bin on my path, as well as Python's home directory. I do not have PYTHONPATH or PYTHONHOME set, and I know I have the correct privileges to see everything.
I have successfully installed pandas for a Windows 32 bit version Python 3.4 with pre-complied code (no Visual Studio required) using the website:
http://www.lfd.uci.edu/~gohlke/pythonlibs/#pyyaml
There is link for Python 3.5 pre-complied code at this site as well, but I have not tested it.
Download the code you want to a directory on your machine.
Using your Windows CMD.exe, go to your python directory and enter:
Python -w pip install "YourDirectory/pandas-0.18.1-cp35-cp35m-win32.whl"
OR
Python -w pip install "YourDirectory/pandas-0.18.1-cp35-cp35m-win_amd64.whl
Choose the version based on the version of Python you have
have, 32 bit or 64 bit.
Good Luck!
Anaconda has included one version of Python with it. You have to change your system environment path with Anaconda's instead of the former one to avoid conflict. Also, if you want to make the whole process easy, it is recommended to use PyCharm, and it will ask you to choose the Python interpreter you want.

ipython for Python 2.7.2 users on PUIAS (RedHat) Linux

I am guessing that I am not the only one using non-system Python 2.7.2 for scientific computations on 6.2 PUIAS Linux (RedHat)). My Python 2.7.2 is installed in the sandbox and I call it with the command python2.7. When I need to execute scripts this is not the problem. However, I would like to use ipython instead default Python shell for interactive use. I added ipython using yum? When I start it, it defaults to system python. How do I force it to load python2.7?
Thanks a bunch!
IPython is a Python package. When you have multiple Pythons, you must (generally, barring PYTHONPATH hacks) install a package for a given Python. yum would only install it for the System Python, so you must install it separately for your own Python. The solution is to simply pip install ipython with your own Python (install distribute/pip first, if you haven't).
I am going to answer my own question. For people who are using Python for scientific computing on RedHat clones (PUIAS for example) the easiest way to get all they need is to use rpm package manager and Enthought Python Distribution (EPD for short). EPD installs everything in a sandbox so system tools which are based on an obsolete version of Python are not massed up. However, paths have to be adjusted for system or even easier on the user base so that the using shell invokes non-system tools.

Categories