___           ___           ___           ___           ___           ___     
           /\  \         /\  \         /\__\         /\  \         /\__\         /\__\    
           \:\  \       /::\  \       /::|  |       /::\  \       /::|  |       /::|  |   
            \:\  \     /:/\:\  \     /:|:|  |      /:/\:\  \     /:|:|  |      /:|:|  |   
            /::\  \   /::\~\:\  \   /:/|:|__|__   /::\~\:\  \   /:/|:|  |__   /:/|:|  |__ 
           /:/\:\__\ /:/\:\ \:\__\ /:/ |::::\__\ /:/\:\ \:\__\ /:/ |:| /\__\ /:/ |:| /\__\
          /:/  \/__/ \:\~\:\ \/__/ \/__/~~/:/  / \/__\:\/:/  / \/__|:|/:/  / \/__|:|/:/  /
         /:/  /       \:\ \:\__\         /:/  /       \::/  /      |:/:/  /      |:/:/  / 
         \/__/         \:\ \/__/        /:/  /        /:/  /       |::/  /       |::/  /  
                        \:\__\         /:/  /        /:/  /        /:/  /        /:/  /   
                         \/__/         \/__/         \/__/         \/__/         \/__/    

TEMANN

• ThermoElectric Materials Artifical Neural Network (TEMANN) is a python package that can be used to predict Seebeck coefficients for novel materials in units of uV/K. All that is required for prediction is the material's chemical formula, the space group of the material, and the temperature (K) of interest.

Use Cases

1. Pipeline for easy datacleaning to accomidate growing dataset for improved ANN training.
   • Supported by the DataSet component.
2. Input novel materials to generate predicted Seebeck coefficent.
   • Supported by the prediction, interpret, query, spacegroup, and util components.
3. Input three elements and generate a ternary heatmap of the Seebeck coefficients.
   • Support by the plotting, interpret, and prediction components.

Example

>>> import temann
>>> temann.predict_seebeck('CaMnO3', 62, 400)
-435.9079284667969
>>> temann.plot_ternary('CaMnO')

Installation

git clone https://github.com/Luochenghuang/TEMANN.git
cd TEMANN
python setup.py install

Workflow

Poster