When reading the TensorNet paper (TensorNet: Cartesian Tensor Representations for Efficient Learning of Molecular Potentials), specifically in section A.3, the authors describe the details of training on scalar, vector, and tensor properties of ethanol.
It turns out that a special output model is required to learn dipole and tensor properties. However, I couldn’t find the corresponding code (output model for dipole and tensor properties) in the repository. Could you please provide the code or point me to where it is located, so I can reproduce the results from the paper?
@guillemsimeon, could you share some details about how you went about this? I do not think those output models are currently present in the latest release.
When reading the TensorNet paper (TensorNet: Cartesian Tensor Representations for Efficient Learning of Molecular Potentials), specifically in section A.3, the authors describe the details of training on scalar, vector, and tensor properties of ethanol.
It turns out that a special output model is required to learn dipole and tensor properties. However, I couldn’t find the corresponding code (output model for dipole and tensor properties) in the repository. Could you please provide the code or point me to where it is located, so I can reproduce the results from the paper?