Develop Reference

Cannot uninstall 'numpy'

I've been trying to install astropy and at the end of the installation I get this message:
Cannot uninstall 'numpy'. It is a distutils installed project and thus
we cannot accurately determine which files belong to it which would
lead to only a partial uninstall.
I have tried: pip uninstall numpy and then I get the same message.
I have Python 2.7.10 in a macOS High Sierra version 13.10.5

This doesn't directly answer your question, but that's because you're asking the wrong question.
Astropy requires Python 3.5 or 3.6. Trying to get it working with Apple's pre-installed Python 2.7 is a waste of time. You might be able to get an old version working this way, but not by using the installation instructions on astropy.org, and it won't be supported even if you do.
The easy solution is to just Install the latest Anaconda 5.x with Python 3.6, because it comes with Astropy built in.
The almost-as-easy solution is to install Python 3.6 from either a python.org binary installer, or Homebrew, and then use pip3 or, better, python3 -m pip to install everything, as explained on the Astropy install page.
Either way, before doing anything else, you want to get back to a clean slate. In particular, you do not want pip, or any other scripts, attached to Apple's Python 2.7; they will only cause confusion. Assuming you can't reinstall macOS from scratch, the best way to do this is:
Look in /Library/Python/2.7/site-packages and delete everything there except for README and Extras.pth.
Look in /usr/local/bin for symlinks to anything in that site-packages. (If you don't know much about using Unix, try this command: ls -l /usr/local/bin | grep 2.7.) You'll probably have pip and pip2.7 here, and probably nothing else. But whatever you have here, delete it.
Now, when you install Python 3.6, the only thing named pip anywhere will be that Python 3.6's pip. You still want to use pip3 or python3 -m pip, but if you screw up and type pip by accident, it won't break anything.
Also, you should strongly consider using a virtual environment. See the Python Packaging Authority's User Guide (or the Anaconda docs, if you went that way) for more on this.

One simple solution I found:
sudo -H pip install astropy --ignore-installed numpy

I also had this issue and couldn't get around it in a clean way, but this is what I did:
Inside the Lib folder search "numpy" for an egg_info file (eg. numpy-1.11.0.dev0_2329eae.egg-info).
In my case, this is what Pip was looking at to determine if a current version already exists. I deleted it, then ran normal
pip install numpy
and installed the newest version.
I don't recommend this because I don't understand what it's doing under the hood and it doesn't properly uninstall the old version which could be a recipe for trouble down the line, but if you're desperate like I was then maybe this is a solution for you.

How to install matplotlib for particular version of Python?

I have installed scipy and numpy, and they are being used with my current, desired version of python 2.7.6 (I am running on OSX Mavericks and had to upgrade.) However, when I pip installed matplotlib, by default it referenced my previous python version, 2.7.5, thus making it troublesome to use (obviously.)
How do I change which version of python matplotlib uses so I can import and use the library?
Thanks.

The way I would solve this problem is like this, firstly one would need to go into your 2.7.6 directory, and under the Scripts folder you will find the pip executable. My suggestion is (because its difficult to debug this kind of people without having all the details) is this:
./pip install matplotlib
And see if this succeeds, otherwise, I suggest using pyenv to manage your python installations.

I suggest you use Macports for installing additional Python versions on OS X. Once Macports is installed, it's fairly easy to install Python 2.7.6. All you'd have to do is:
sudo port install python27
Now, you should be able to get all the libraries you need just as easily, using, too, Macports.
sudo port install py27-numpy
sudo port install py27-scipy
sudo port install py27-matplotlib
Macports should solve all the dependencies and, of course, link the packages to their correct Python versions, avoiding you a lot of headaches.
For a step by step guide on how to set up a nice, functional Python environment, visit: http://jakevdp.github.io/blog/2013/02/02/setting-up-a-mac-for-python-development/

installing modules in python - pip, distribute, nose, virtualenv

I'm aware that there are similar questions on SO. This one, for example: What's the proper way to install pip, virtualenv, and distribute for Python?
I'd like to install these modules as per my Learn Python the Hard Way tutorial: http://learnpythonthehardway.org/book/ex46.html
I managed (I think) to install pip by using sudo easy_install pip but when I then ran pydoc modules I could not see it. So I'm not even sure it's installed.
The answer above in question 4324558 is difficult for me to understand: what's a bootstrap, what's curl and why would I set up a virtual environment? Yes, as a learner I should try to pick up as much as I can but I don't want to first create the universe, I just want to get the task at hand done.
How do I install these modules? Is it as complicated as it sounds in the quoted answer? The top voted answer says "Install virtualenv into a bootstrap virtual environment. Use the that virtual environment to create more. Since virtualenv ships with pip and distribute, you get everything from one install."
I really don't get what all that means. Isn't there something about the "Zen" of python and a one true way to get things done? Or am I out of context here? What is "the right way" to install these modules?
I tried:
pip install virtualenv in the terminal and received the following output:
Wheel installs require setuptools >= 0.8 for dist-info support.
pip's wheel support requires setuptools >= 0.8 for dist-info support.
Storing debug log for failure in /Users/myname/.pip/pip.log
I'm using a Mac and python 2.7

To solve your issue,
Just install (or upgrade) the setuptools:
sudo easy_install -U setuptools
Then you can run again: pip install virtualenv

Try adding 'sudo' in your command as-
sudo pip install virtualenv
It worked for me.

Have a look at Python Development Environment on Mac OS X Mavericks 10.9.
I followed these steps as well when trying to get Python 2.7 and Python 3.3 installed on OS X. It doesn't tell you how to install nose and distribute, but you should have a working environment and you can pick up from there.
I did have a problem using virtualenv and pip with Python 3, the question and solutions is available here.

Import Error: No module named AppKit

I use Mac OS X Lion and Python 2.7. I am new to python. Can anyone tell me how to import AppKit and PyObjC to Python. But i get the errors when trying to import Import Error: No module named AppKit or 'Import Error: No module named PyObjC.
Trying easy_install does not help either.
What can i do to import these 2 modules?

If not already installed, install pip by running:
sudo easy_install pip
Then run:
## install for all users
sudo pip install pyobjc
or
## install for current user only
pip install pyobjc --user
NOTE: the general recommendation is to avoid using the system python, and use instead a user-maintained version, for example installed via brew install python3, macports, conda or whatever you already use for third party dependencies.

If you're on Mac it could be you're using the incorrect pip version, try sudo pip3 install -U pyobjc that solved it for me.

This a good guide on how to install PyObjc: http://ioanna.me/2009/09/installing-pyobjc-xcode-templates-in-snow-leopard/
When I was trying to install PyObjc I found it quite hard, but after a while, (and lots of google searches) I got it working.
Edit
I've found a more modern guide on how to do this. Download the templates and put them into your ~/Library/Developer/Xcode/Templates/ directory, and everything should work correctly.

If you are using PyCharm, you should install PyObjC package manually within it because of venv I guess. This was the only working solution for me.

In my case none of them worked. I was using Anaconda in my mac, later I discovered that the main issue with the Anaconda. Then I installed PyObjC with it's all supporting libraries inside Anaconda in a specific Environment.
I run the command line by clicking the Environment play icon and run my script.
It worked without any issues :)

Python cannot find dateutil.relativedelta

I am trying to run a program using paster serve, but I keep getting the error:
ImportError: No module named dateutil.relativedelta
I am running Python version 2.6.7 and dateutil version 1.5, so it should be installed.
Has anyone got any ideas as to why this would happen?
I am importing using
from dateutil.relativedelta import *
I can even see the package when I search:
/usr/lib/python2.7/site-packages/dateutil/relativedelta.pyc
/usr/lib/python2.7/site-packages/dateutil/relativedelta.py
/usr/lib/python2.7/site-packages/dateutil/relativedelta.pyo
UPDATE
Immediately I look at this and see that dateutil is only installed for Python 2.7, and I bet what I was doing was this:
sudo yum install python-dateutil
To which sudo would have switch to the default Python version (i.e., Python 2.7 instead of 2.6.4).
Solving this would have been as simple as:
su
(switch to virtual environment)
yum install python-dateutil
Using su and then switching to the virtual environment will give root access and install to the virtual Python directory. Using sudo will install libraries to the default directory, not the virtual environments site-packages.

I also ran into this issue. The simple solution I ended up using was to add --upgrade to the end of the command. This forced it to install it even though Python thought it was installed. This resolved the issue.
So if you have this issue, try the following:
sudo pip install python-dateutil --upgrade
It can't possibly hurt anything, so there is no harm in just forcing it to be reinstalled.

I had a similar issue but for a simpler reason. My fresh virtualenv simply didn't have dateutil installed and I didn't know the Python package name. I tried pip install dateutil, which obviously didn't work since the package name was incorrect. Running pip install python-dateutil instead worked (without resorting to sudo).

This looks like a problem of package installation to me. A troubleshooting list that comes to my mind:
Verify you installed the package.
If installed, verify that the files have been stored in the right directory (a directory accessible from your python interpreter (= in the PYTHONPATH, useful article here).
Verify permission on those files.
Restart your shell if you tried the import there.
Reboot your computer (ouch... it's 10 years since I started using GNU/Linux, but I still suffer from the bad memories of Windows! ;)

(The previous comment about installing python-dateutil helped me, so perhaps my comment helps someone else).
For those on Mac OS (v10.6 (Snow Leopard); I am not sure about other versions), the dateutils package is located by default at:
/System/Library/Frameworks/Python.framework/Versions/2.6/Extras/lib/python/dateutil
whereas pip install writes the package out to:
/Library/Python/2.6/site-packages
and does not update the /Library/Python/2.6/site-packages/easy-install.pth file. As a result, when you import dateutil, you will still point to the old location, you can verify this by "import dateutil; dateutil.__file__".
So what I did (probably better methods are available) was to rename the old directory (/System/Library/.../dateutil) to dateutil.obsolete and restarted Python, then ran the same set of commands again. This doesn't do anything to the path file or sys.path, but skips the old dateutils package so you can get to the new one.