The Anaconda scientific Python distribution by Continuum Analytics is recommended for Windows users. You can use pycalphad with Python 2 or Python 3, but we recommend Python 3 for the best experience. After you have installed either Anaconda or Miniconda, use conda config --add channels richardotis conda-forge followed by conda install pycalphad to install. Note that you will need to have a working C/C++ compiler for pycalphad to work, so conda install mingw may also be necessary on Windows. To install the package into an isolated environment, use conda create -c richardotis -n [envname] pycalphad Then use source activate [envname] on Linux/OSX or activate [envname] on Windows to enter the environment.
For interactive pycalphad sessions, we recommend installing the Jupyter Notebook.
If not using a special distribution like Canopy or Anaconda, it's recommended to install pycalphad in a virtualenv using virtualenvwrapper. pip install pycalphad inside the virtualenv will install with any required dependencies. You may also want to pip install fastcache for a mild performance boost. If you are using Anaconda, see the Windows instructions.
If not using a special distribution like Canopy or Anaconda, it's recommended to install pycalphad in a virtualenv using virtualenvwrapper. pip install pycalphad inside the virtualenv will install with any required dependencies. You may also want to pip install fastcache for a mild performance boost. If you are using Anaconda, see the Windows instructions.
git clone https://github.com/pycalphad/pycalphad.git pycalphad/- Using conda:
conda config --add channels richardotis conda-forgeconda create -n [envname] pycalphadconda develop -n [envname] pycalphad/source activate [envname]on Linux/OSX oractivate [envname]on Windows to enter the environment.
- Or, inside a virtualenv:
python setup.py develop