# Installing Halotools using a virtual environment¶

If you use conda to manage your python distribution and package dependencies, it is easy to create a virtual environment that will automatically have compatible versions of the necessary dependencies required by Halotools. By installing into a virtual environment, you will not change any of the packages that are already installed system-wide on your machine. In the example below, we will use conda to create a virtual environment with all the dependencies handled automatically:

conda create -n halotools_env astropy numpy scipy h5py requests beautifulsoup4 cython python=2.7.11


In order to activate this environment:

source activate halotools_env


Then install halotools into this environment:

pip install halotools


Any additional packages you install into the halotools_env virtual environment will not impact your system-wide environment. Whenever you want to do science involving Halotools, just activate the environment and import the code. When you are done and wish to return to your normal system environment:

source deactivate