PySIT has the following dependencies

- Python 3.6 (or greater; So far we still need Python 2.7 to install the toolbox ––– PETSC)

- NumPy 1.15 (or greater)

- SciPy 1.1 (or greater)

- matplotlib 2.2.3 (or greater)

- PyAMG 4.0.0 (or greater)

- ObsPy 1.1 (or greater)

- PETSC4Py 3.6 (for MAC OS) and PETSC4Py 3.9 (for Linux)

External packages:

- PETSC 3.6 (for MAC OS) and PETSC 3.9 (for Linux)

For optional parallel support, PySIT can depend on

- MPI4Py 3.0.0

On all platforms (Linux, Windows 7 or greater, and MAC OS X), we recommend a preassembled scientific python distribution, such as Continuum IO’s Anaconda or Enthought’s Canopy. These collections already include compatible (and in some cases accelerated) versions of most of PySIT’s dependencies. Download and follow the appropriate instructions for your operating system/distribution. In this instruction, we will show a step by step example that uses Miniconda3 to install PySIT.

1. The easiest way to install Miniconda3 is to download it from the download page of Miniconda. Please select the Miniconda that works for your platform. There are two versions Python 3.7 and Python 2.7. Please download the one with Python 3.7 that corresponds to Miniconda3, while the one with Python 2.7 corresponds to Python 2.7.

2. Download the proper installation file to your local directory, for example ~/Download (In this example, we assume that the installation file of miniconda is located at ~/Download. You can change it to any directory that you want). You will find a file named similar as Miniconda3-latest-MacOSX-x86_64.sh (for MAC OS users) or Miniconda3-latest-Linux-x86_64.sh (for Linux users) in your directory.

3. Open your terminal. Go to the directory ~/Download by executing the following command:

   $ cd ~/Download

   Then, install Miniconda3 by executing the following command for MAC OS users:

   $ source ./Miniconda3-latest-MacOSX-x86_64.sh

   and executing the following command for Linux useres:

   $ source ./Miniconda3-latest-Linux-x86_64.sh

   After the installation, you can check if Miniconda3 has been installed successfully by executing the command:

   $ which conda

   If it has been installed, then you will see the following output on the screen:

   $ /YOURHOMEDIRECTORY/miniconda3/bin/conda
   

1. Create a Python3 virtual environment named with myenv with necessary packages by excuting the following command

   $ conda create -n myenv numpy=1.15.1 scipy=1.1.0 matplotlib=2.2.3 pyamg=4.0.0 

   This command will create a virtual Python environment with most of the dependencies.

2. Activate your environment by the following command:

   $ source activate myenv

   If the environment is created successfully, you would see the following sentence at the left bottom of your terminal:

   $(myenv)YourComputerName:

   where YourComputerName stands for the actual name of your computer.

3. Install ObsPy by pip. We assume that you already have pip installed. If not, please go to the webpage of pip to install pip first. With pip installed, you can install ObsPy by executing the following command:

   $ pip install obspy

4. Install necessary external softwares including PETSC and MUMPS. Assume that you clone or download the PySIT package at directory /PATHTOPYSIT. First go to the directory by executing the following command:

   $ cd /PATHTOPYSIT

   Then, execute the following commands to install PETSC and MUMPS. For linux users please use:

   $ source ./install_petsc4py_linux.sh

   For MAC OS users:

   $ source ./install_petsc4py_OSx.sh
   

5. Install the PySIT toolbox. First, make sure that you are in the PySIT directory by executing the following command:

   $ cd /PATHTOPYSIT

   Then, execute the following command:

   $ python setup.py install

6. Check if PySIT has been successfully installed. Please open a python by

   $ python

   Then try to import the PySIT toolbox by

   $ import pysit

   If there are no warnings or errors on the screen, then congratulations, you have successfully installed the PySIT toolbox. Please feel free to work with this powerful toolbox from the examples.