Published February 2, 2021 | Version 1.0
Dataset Open

Optical absorption of metal oxides and results of machine learning predictions

Gregoire, John M1 ORCID icon
Kong, Shufeng2
Guevarra, Dan1
Gomes, Carla P.2
  • 1. ROR icon California Institute of Technology
  • 2. ROR icon Cornell University

Citation

Gregoire, J. M., Kong, S., Guevarra, D., & Gomes, C. P. (2021). Optical absorption of metal oxides and results of machine learning predictions (1.0) [Data set]. CaltechDATA. https://doi.org/10.22002/D1.1878

Description

CSV files that collectively contain optical absorption data from metal oxides with different cation elements compositions. The optical properties for each metal oxide composition are the 10-dimensional unitless absorption coefficient where the 10 dimensions correspond to equally-spaced ranges of photon energy spanning 1.39 eV to 3.11 eV. These data also include test-train splits for benchmarking machine learning models that predict the optical absorption, along with results obtained to-date with various machine learning models. For each test-train split, data are normalized to 0 mean, unit variance.

Files

results_69ternaries.csv
Files (88.7 MB)
Name Size
results_69ternaries.csv
md5:eb589181111a360fbf099a08444e73ed
10.3 MB Preview Download
Predict_3cation_testtrain_69ternaries.csv
md5:260163fca7e387d33093693e93139ab2
43.1 MB Preview Download
prediction_new_ternary_compositions_hclmpt.csv
md5:20bd5a2c8acd3cb7d67772e68d293d94
34.5 MB Preview Download
readme.txt
md5:4d21b481dbc6874615c83d5cea2b0b58
1.8 kB Preview Download
results_random_split.csv
md5:b9f3aba0fe35415bd3dcc26c8418c7ce
791.5 kB Preview Download

Additional details

Identifiers Funding
Created:
September 8, 2022
Modified:
November 18, 2022