Installation with Anaconda/Miniconda
[1] Download and install Anaconda (or Miniconda if you prefer). Use the Python 3.X version if you have no preference, but either will do.
[2] Open an Anaconda prompt window (Windows) or the Terminal (Mac/Linux).
[3] Create a new environment for Pytzer:
conda create -n pytzer python=3.7.3 numpy=1.16.1 scipy=1.2.1
Other recent versions of Python, NumPy and SciPy will probably work fine too, but they have not been tested with Pytzer.
[4] Activate the new environment:
conda activate pytzer # Windows or Linux
source activate pytzer # Mac
You should now see the name of the environment (i.e. pytzer) appear at the start of each new line in the command window.
[5] Install Pytzer into the pytzer environment:
pip install pytzer
pip install git+https://github.com/mvdh7/autograd#egg=autograd --upgrade --no-cache-dir
The second line above is strongly recommended, but optional. It upgrades Autograd to the latest version that has been tested with Pytzer, which eliminates some deprecation warnings that may appear when using the relatively old Autograd version available from PyPI. You could also switch mvdh7 in the URL to HIPS to get the very latest Autograd straight from the horse's mouth.
Running Pytzer as a "black box"
You can just provide Pytzer with a CSV file of temperature, pressures and molality values, and have it return the corresponding activity coefficients in a new CSV file, only having to write the bare minimum in Python, as follows.
[1] Create an input CSV file as described in the import/export documentation - or just save and use the example file pytzerQuickStart.csv.
[2] Open an Anaconda prompt window (Windows) or the Terminal (Mac/Linux) and activate the pytzer environment (i.e. installation step [4]).
[3] Navigate to the folder containing the input CSV file (using cd).
[4] Start Python:
python
[5] Import Pytzer (as pz, by convention) and then run its "black box" function on your input file:
>>> import pytzer as pz
>>> pz.blackbox('pytzerQuickStart.csv')
Once the calculations are complete, a new CSV file will appear, in the same folder as the input file, with the same name as the input file but with _py appended. It contains the input temperature and molality values, followed by the osmotic coefficient (column header: osm), water activity (aw), and then the activity coefficient of each solute (with column headers e.g. gNa for the $\gamma(\ce{Na+})$ activity coefficient).
The black box function is currently set up to use the Seawater coefficient library. This is based on the model of Waters and Millero [WM13], with a few modifications. It is still under development, so the results will probably change as Pytzer is updated. The Seawater coefficient library contains coefficients for the components: $\ce{Na+}$, $\ce{K+}$, $\ce{Ca^2+}$, $\ce{Mg^2+}$, $\ce{MgOH+}$, $\ce{H+}$, $\ce{OH-}$, $\ce{Cl-}$, $\ce{HSO4-}$, $\ce{SO4^2-}$, $\ce{tris}$, and $\ce{trisH+}$. If you have any other solutes in your input file, Pytzer should still work (as long the charges on those solutes are defined in properties.charges), but all interaction coefficients involving those solutes will be set to zero.
Updating Pytzer
[1] Open an Anaconda prompt window (Windows) or the Terminal (Mac/Linux) and activate the pytzer environment (installation step [4]).
[2] Update Pytzer with pip:
pip install pytzer --upgrade --no-cache-dir