Action: PYTORCH_MODEL

Module	pytorch
Description	Usage
Load a PyTorch model compiled with TorchScript.

Details and examples

Load a PyTorch model compiled with TorchScript.

This can be a function defined in Python or a more complex model, such as a neural network optimized on a set of data. In both cases the derivatives of the outputs with respect to the inputs are computed using the automatic differentiation (autograd) feature of Pytorch.

By default it is assumed that the model is saved as: model.ptc, unless otherwise indicated by the FILE keyword. The function automatically checks for the number of output dimensions and creates a component for each of them. The outputs are called node-i with i between 0 and N-1 for N outputs.

Note that this function requires an installation of the LibTorch C++ library. Check the instructions on the module page for instructions on how to enable the module.

Examples

Load a model called torch_model.ptc that takes as input two dihedral angles and returns two outputs.

Click on the labels of the actions for more information on what each action computes

phiThe TORSION action with label phi calculates the following quantities:
 Quantity   
 Type   
 Description  
phi
scalar
the TORSION involving these atoms: TORSIONCalculate one or multiple torsional angles. More details ATOMSthe four atoms involved in the torsional angle=5,7,9,15
psiThe TORSION action with label psi calculates the following quantities:
 Quantity   
 Type   
 Description  
psi
scalar
the TORSION involving these atoms: TORSIONCalculate one or multiple torsional angles. More details ATOMSthe four atoms involved in the torsional angle=7,9,15,17
modelThe PYTORCH_MODEL action with label model calculates the following quantities:
 Quantity   
 Type   
 Description  
model.node-0
scalar
Model outputs  This is the 0th of these quantities
model.node-1
scalar
Model outputs  This is the 1th of these quantities: PYTORCH_MODELLoad a PyTorch model compiled with TorchScript. More details ...
  FILEFilename of the PyTorch compiled model=regtest/pytorch/rt-pytorch_model_2d/torch_model.ptc
  ARGthe labels of the values from which the function is calculated=phi,psi
...
PRINTPrint quantities to a file. More details FILEthe name of the file on which to output these quantities=COLVAR ARGthe labels of the values that you would like to print to the file=model.node-0,model.node-1

Input

The arguments that serve as the input for this action are specified using one or more of the keywords in the following table.

Keyword	Type	Description
ARG	scalar	the labels of the values from which the function is calculated

Full list of keywords

The following table describes the keywords and options that can be used with this action

Keyword	Type	Default	Description
ARG	input	none	the labels of the values from which the function is calculated
NUMERICAL_DERIVATIVESThis keyword do not have examples	optional	false	calculate the derivatives for these quantities numerically
FILE	optional	not used	Filename of the PyTorch compiled model