ANIMA

[!WARNING]
This repo was moved to Github at https://github.com/rpcarvs/anima

ANIMA (AdvaNced artificial Intelligence for Materials reseArch) is a machine learning code developed to work with Quantum Mechanics and Materials Science. For now, the main goal is related to energy materials and organic electroactive materials. The software includes a collection of tools and functions to process data in a molecular or solid-state level and uses the processed data to build a ML model based on Neural Networks. The user has the freedom to choose between our NN (see "models" class) or any external framework, e.g., Tensorflow/PyTorch.

Here are a few representations for the molecular/crystals systems available: * Couloumb Matrix * Many Body Tensor Representation * SMILES

The SMILES package also contains a set of text-based functions in order to prepare the SMILES string for a natural language processing (NLP) approach, if necessary.

Dependencies:

This package is based on common python libraries, except in specific cases. * numpy * scipy * pandas * joblib * sklearn (only used for a simple scaling and will be removed in a future version)

ONLY FOR THE SMILES CLASS: * rdkit * openbabel/pybel (version 3 or later is recommended) * pysmiles

ONLY FOR THE AI-kernel model (in models module): * pytorch (model was developed using version 1.7)

It is also recommended to install the dependencies in a new python-environment and using Anaconda.

Please, feel free to contact me at rodrigo.carvalho.al@gmail.com, rodrigo.carvalho@physics.uu.se or rodrigo.carvalho@kemi.uu.se