Getting Started with SimPEG#
Prerequisite: Installing Python#
SimPEG is written in Python! We highly recommend installing it using Anaconda (or the alternative Miniforge). It installs Python, Jupyter and other core Python libraries for scientific computing. If you and Python are not yet acquainted, we highly recommend checking out Software Carpentry.
Note
As of version 0.11.0, we will no longer ensure compatibility with Python 2.7. Please use the latest version of Python 3 with SimPEG. For more information on the transition of the Python ecosystem to Python 3, please see the Python 3 Statement.
Installing SimPEG#
Conda Forge#
SimPEG is available through conda-forge and you can install is using the conda package manager that comes with the Anaconda or Miniforge distributions:
conda install SimPEG --channel conda-forge
Installing through conda is our recommended method of installation.
Note
Since version 23.10.0,
conda
makes use of the libmamba
solver to resolve dependencies. It
makes creation of environments and installation of new packages much faster
than when using older versions of conda
.
Since this version, conda
can achieve the same performance as
mamba
, so there’s no need to install mamba
if you have an updated
version of conda
.
If not, either update conda, or
keep using mamba
instead.
PyPi#
SimPEG is on pypi! First, make sure your version of pip is up-to-date
pip install --upgrade pip
Then you can install SimPEG
pip install SimPEG
To update SimPEG, you can run
pip install --upgrade SimPEG
Installing from Source#
First (you need git):
git clone https://github.com/simpeg/simpeg
Second (from the root of the SimPEG repository):
pip install .
If you are interested in contributing to SimPEG, please check out the page on Contributing
Success?#
If you have been successful at downloading and installing SimPEG, you should be able to download and run any of the examples and tutorials.
If not, you can reach out to other people developing and using SimPEG on the google forum or on Mattermost.
Useful Links#
An enormous amount of information (including tutorials and examples) can be found on the official websites of the packages
Python for scientific computing#
Python for Scientists Links to commonly used packages, Matlab to Python comparison
Python Wiki Lists packages and resources for scientific computing in Python
Numpy and Matlab#
Lessons in Python#
Editing Python#
There are numerous ways to edit and test Python (see PythonWiki for an overview) and in our group at least the following options are being used: