Yokoyama et al., Supplementary Data for "Prediction of Li-ion Conductivity in Ca and Si co-doped LiZr₂(PO₄)₃ Using a Denoising Autoencoder for Experimental Data"

uniDAE.py is the python script used in this study. The script includes a denoising autoencoder for XRD profiles with six attention heads and a deep learning model for regression analysis of the activation energy (Ea) of ion conduction. bestmodel.pth is the trained deep learning model for uniDAE.py used in this study. Input folder contains csv files that is training/validation/test data used in uniDAE.py. (LCZSP_data_with_Ea.csv includes sample number, heating condition, and Ea data. LCZSP_XRD_original_withH.csv contains the same information as LCZSP_data.zip (the Ea values at each sintering temperature). It is formatted for input into the machine learning program. LCZSP_data.zip (contains the same information of LCZSP_XRD_original_withH.csv)Experimental measurement data for the lithium-ion conductor Li1.45Ca0.15Zr1.85Si0.15P2.85O12 (LCZSP) material synthesized using 54 different sintering programs. Reference Yokoyama et al. DOI: 10.1063/5.0231411 For information on sample synthesis (LCZSP_data.zip), refer to: H. Takeda, H. Fukuda, K. Nakano, S. Hashimura, N. Tanibata, M. Nakayama, Y. Ono, T. Natori, "Process optimisation for NASICON-type solid electrolyte synthesis using a combination of experiments and Bayesian optimisation", Mater. Adv., 3, 8141-8148. (2022) DOI: 10.1039/D2MA00731B.