{"contributor-id" "729049db-685a-43b1-97a8-617daa2586ba"
 "description" "A parallel MACE equivariant graph neural network for Si. This is a general purpose MLIP for Silicon, which can be employed for miscellaneous tasks owing to the broad dataset used in training. It was trained with a cutoff radius of 4 angstroms with max order of spherical harmonics set to 1. The hidden representation was set to 32x0e + 32x1o, meaning that it has equivariant symmetry equivalent to spherical harmonics of order 1 (1o). Number of graph convolutions to 3, correlation order was set to 2 (max tensor product body order = 3). For training, the GAP Si PRX (Bartók et al. Phys. Rev. X 8, 041048) dataset was used. The model was trained on energy and forces, with weight 1 and 1000 respectively. The model was trained till convergence, then the best model performing model on validation set selected."
 "developer" ["729049db-685a-43b1-97a8-617daa2586ba"
              "e1b97ecf-68df-423a-97de-e11a40dc4dde"
              "360c0aed-48ce-45f6-ba13-337f12a531e8"]
 "doi" "10.25950/6f365f7b"
 "domain" "openkim.org"
 "executables" []
 "extended-id" "TorchML_MACE_GuptaTadmorMartiniani_2024_Si__MO_781946209112_000"
 "kim-api-version" "2.3"
 "maintainer-id" "729049db-685a-43b1-97a8-617daa2586ba"
 "model-driver" "TorchML__MD_173118614730_000"
 "potential-type" "mace"
 "publication-year" "2024"
 "source-citations" [{"author" "Bart{\\'o}k, Albert P and Kermode, James and Bernstein, Noam and Cs{\\'a}nyi, G{\\'a}bor"
                      "doi" "10.1103/PhysRevX.8.041048"
                      "issue" "4"
                      "journal" "Physical Review X"
                      "pages" "041048"
                      "publisher" "American Physical Society"
                      "recordkey" "MO_196181738937_000a"
                      "recordtype" "article"
                      "title" "Machine learning a general-purpose interatomic potential for silicon"
                      "volume" "8"
                      "year" "2018"}]
 "species" ["Si"]
 "title" "Parallel MACE Equivariant GNN for Si developed by Gupta et al. (2024) v000"}