The steps below are for a non-administrative install.

If you want to install and maintain Anaconda for all users, please go to this version of the instructions. (Admin access required)

What is the Anaconda distribution?

Anaconda is an open-source Python distribution that makes is possible to easily install and manage many pre-packaged third party Python modules. It is a competing fully open-source alternative to the Enthought Python Distribution (EPD).

It has some big advantages over using the Python Esri provides with ArcGIS:

Workflow

The general workflow to make this happen is to:

1) Install Anaconda (Single-User - no admin access needed)

  1. Go to http://continuum.io/downloads
  2. Download the 32-bit (x86) version of Anaconda (ArcMap [32-bit] can only directly work with 32-bit Python)
  3. In the install dialogs:
    1. Select install for a single user (Not "All Users")
    2. Install to a folder where there is going to be plenty of space (recommend the D drive, not the C drive)
    3. IMPORTANT: To avoid breaking ArcGIS, uncheck the checkboxes  (a) make Anaconda the default Python and (b) add Anaconda's Python to the PATH. 
    4. Here are screenshots to help guide you through the install process.

 

Python 2.7.11 has a registry conflict with ArcGIS Desktop 10.3. This must be fixed before you continue:

Esri KB 45924 http://support.esri.com/en/knowledgebase/techarticles/detail/45924

 

At this point, you will have full Anaconda version of the python environment that's spiffy and new and totally useless with ArcMap. 

2) Configure Anaconda To Work with ArcGIS

The following workflow will demonstrate how set up a custom Python environment within Anaconda that is similar to that of ArcGIS Python, and then add a compatible version pandas  module. Pandas is great for a whole variety of data processing tasks (just google "python pandas"). If you get this far, you can adapt this workflow to include any other modules you may want.

The following example is for ArcGIS 10.2.2.

A. Find the versions of numpy and matplotlib ArcGIS is using.

Open ArcMap and its Python window, and enter these commands:

>>> import sys, numpy, matplotlib
>>> print(sys.version, numpy.__version__, matplotlib.__version__)
'2.7.5 (default, May 15 2013, 22:43:36) [MSC v.1500 32 bit (Intel)]', '1.7.1', '1.3.0')

B. Create a custom Anaconda environment (including pandas) that is compatible with ArcGIS 10.2.2.

The critical modules for compatibility are the three above. We'll also include a few others that we know are shipped the ArcGIS 10.2.2's Python stack.

  1. Get to the Anaconda Command Prompt (Start > Search > Anaconda, pick "Anaconda Command Prompt")
  2. Type (depending on Arc version):
    ArcGIS 10.2.2:  "conda create -n arc1022 python=2.7.5 numpy=1.7.1 matplotlib=1.3.0 pyparsing xlrd xlwt" 
    ArcGIS 10.3.1:  "conda create -n arc1031 python=2.7.8 numpy=1.7.1 matplotlib=1.3.0 pyparsing xlrd xlwt"
    ArcGIS 10.4: "conda create -n arc104 python=2.7.10 numpy=1.9.2 matplotlib=1.4.3 pyparsing xlrd xlwt"
  3. The conda tool will:

C. Test the virtual environment

Since this new custom environment is not the Anaconda default, you need to let it know that's what you want to use. Then, you'll ask it to tell you what's installed. So, at the Anaconda Command Prompt, type:

 

D:\Users\cprice>activate arc1022
Activating environment "arc1022"...
[arc1022] D:\Users\cprice>conda list
# packages in environment at D:\Users\cprice\Anaconda\envs\arc1022:
#
dateutil 2.4.1 py27_0
matplotlib 1.3.0
np17py27_0numpy 1.7.1 py27_3
...
[arc1022] D:\Users\cprice>python
Python 2.7.5 |Continuum Analytics, Inc.| (default, Jul 1 2013, 12:41:55) [MSC v.1500 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>>

D. Add more packages

You can add more packages using conda install, but make sure you specify version numbers for these that won't change the environment's version of python or numpy (or ArcGIS will not be able to use that environment anymore).

Let's add the pandas module.

Specifying python, numpy, and matplotlib versions makes sure the environment will still work with ArcGIS, and Anaconda determines the most recent compatible version of pandas (in our case, 0.13.0).

conda install -n arc1022 python=2.7.5 matplotlib=1.3.0 numpy=1.7.1 pandas
...
The following NEW packages will be INSTALLED:
pandas: 0.13.0-np17py27_0 ### this version works with python 2.7 and numpy 1.7
... 

A few of the most popular things to add to your environment is interactive python ("ipython"), in several flavors:

conda install -n arc1022 python=2.7.5 matplotlib=1.3.0 numpy=1.7.1 ipython ipython-notebook ipython-qtconsole

Here is another example, that installs both pandas and spyder, which includes a whole host of tools (including iPython and iPython notebook).

conda install -n arc1022 python=2.7.5 matplotlib=1.3.0 numpy=1.7.1 pandas spyder

You can search for more packages available in conda with conda search.

You're not limited to adding only packages to which conda has access to your new environment. Here's an example on how to install a commonly desired GIS-related package (shapely) into a conda virtual environment using the python pip utility:

http://deparkes.co.uk/2015/01/29/install-shapely-on-anaconda/

3) Configure ArcGIS to see Anaconda and vice versa

This can easily be done with a Python startup script

  1. Install Anaconda, setup environment to match your ArcGIS version
  2. Edit the paths in this script (usercustomize.py.txt) to match your setup
  3. Put the script in the Python user site packages folder (must be named usercustomize.py)
    The user site-packages folder path can be found with:  python -m site --user-site
    Usually:  C:\Users\username\AppData\Roaming\Python\Python27\site-packages

Testing and Troubleshooting

  1. Start ArcMap, open the Python window
  2. import pandas
  3. If you don't get an error message, you are probably good to go.

If you get an error message when you try to import, a common cause is version incompatibilities between  Esri's python environment modules and the conda modules. (Condo takes care of compatibility across its environment, but it can't know about versions Esri uses in their python install.) In the pandas example, version 0.10.1 also appears to use the same 1.6.1 version of numpy used by ArcGIS 10.1.x, but some errors occur when trying to import pandas due to some other incompatibility. The solution was to modify the Anaconda environment to use an older version of pandas (0.10.0):

  1. Open Anaconda Command Prompt (via the Start menu on Windows, e.g.)
  2. conda install -n arc101 pandas=0.10.0

References

http://www.continuum.io/blog/conda

How to do a separate Python installation with ArcGIS? (GIS Stack Exchange)

Using ArcPy with Anaconda (PyMorton)