Get Started with the Anaconda Quick Start Guide [pdf]
It's worth noting an important difference between Miniconda3 and regular Python 2 Miniconda/Anaconda. With Miniconda3, the default (root) Python will be Python 3. In particular, this means that conda itself will run from Python 3, so when you do "conda create -n python" or "conda create -n package" for some package that depends on Python without specifying a Python version, it will default to Python 3. The Python 2 version of Miniconda/Anaconda would default to Python 3. Of course, in either case, you can override it by passing python=2 or python=3.I personally use Miniconda3 for this reason, because I want Python 3 to be my *default* Python, unless I specify otherwise.
http://repo.continuum.io/miniconda/
| Miniconda3-3.0.5-Windows-x86.exe | 31.9M | 2014-02-18 10:16:34 |
If you're not quite ready to get your feet that wet, stick with Anaconda or Miniconda.
Downloads
Anaconda
Free for all to use: (including commercial)
Linux installers
Linux 32-bit / 411M / md5: f1505963a1c7d2bfe7a73c079b22762dLinux 64-bit / 483M / md5: 9d973e9ac715ce3241c3785704565971
Mac OS X installers
Mac OS X (10.7 or higher - GUI installer) 64-bit / 280M / md5: 772b8e5dc385bf5ea3f78cdd21a8ec71Mac OS X (10.5 or higher - bash installer) 64-bit / 245M / md5: 6ef81bc54a6ab506f352b5589ea80f81
Windows installers
Windows 32-bit / 311M / md5: 5404da4f89dca1a4f5c9efd5ae6fbc5aWindows 64-bit / 367M / md5: b8a404c9f5bfd2452316db3710d2b8ef
Zipped Windows Installers - zipped Windows executable files for those behind firewalls
You can find tutorials, documentation, and other helpful guides for learning and
using Python at http://www.python.org/doc/. Some other resources you may want
to check out: Scipy Lectures, PyData Tutorial videos, 10 Minute Intro to Pandas,
Pandas + Matplotlib plotting tutorial and IPython Notebook-based Lectures on
Learning Python. Also, Continuum Analytics offers Python training classes and a
Free Anaconda Support Forum.
conda
We recommend that you familiarize yourself with conda, an open source package
and environment manager developed by Continuum Analytics. conda is designed to
work very well with the complex binary packages found in the scientific and
numerical Python ecosystem, which are oftentimes very difficult to build and install
using standard Python packaging tools. When a conda package is available for a
particular library, it is super easy to install it into Anaconda. It’s also easy to update
packages, install a particular version of a package, and mix and match packages into
environments with your choice of Python version. To view the conda help menu,
type the command conda -h from the terminal or command line.