The pymetamodels package combines machine learning (ML) metamodeling and analysis tools for the virtual development of modeling systems within a common abstract framework implemented in an accessible and distributable Python package. The development of pymetamodels package is oriented to support ML applications in material science, material informatics and the construction of materials, components and systems soft metamodels informed by hard physics-based modelling (continuum, mesosocopic, ... ) and experimental characterisations.
Basic turtorials and advanced examples can be found in the tutorials section pymetamodels.readthedocs.io.
The package has been build in ITAINNOVA. And is distributed with permissive MIT license.
To install the latest stable version of pymetamodels via pip from PyPI together with all the dependencies, run the following command:
pip install pymetamodels
First steps, basic turtorials an advanced examples can be found in the documentation tutorials section pymetamodels.readthedocs.io. To load and test installation try,
import pymetamodels
### Load main object
mita = pymetamodels.metamodel()
### Load main object (alternative)
mita = pymetamodels.load()
Pymetamodels requires Python >3.7 or an above of release Python.org.
Pymetamodels requires NumPy, SciPy, sklean, matplotlib and SALib installed on your computer. Using pip, these libraries can be installed with the following command:
pip install numpy scipy scikit-learn matplotlib SALib Pillow xlrd xlwt xlutils
The packages are normally included with most Python bundles, such as Anaconda. Generally, they are installed automatically when using pip to install pymetamodels.