OpenFF 2.0.0-rc.1

Contained within this release tarball are the main inputs and outputs used to generate the OpenFF 2.0.0-rc.1
force field, as well as copies of the force field itself.

OpenFF 2.0.0-rc.1 was created using a multi-stage optimization starting from the OpenFF 1.3.0 force field:

1. A select set of the vdW parameters were trained against a set of experimental mass density
   and enthalpy of mixing measurements sourced from the NIST ThermoML archive.

   The DOI of each individual data point can be found in the JSON serialized
   vdw-v1/targets/phys-prop/training-set.json file which can be readily parsed using the
   OpenFF evaluator framework.

   The full inputs and relevant outputs are located in the vdw-v1 directory.

2. A select set of the bond length, bond force constant, equilibrium angle, angle force constant
   and torsion barrier height parameters were trained against a set of QC computed optimized geometries
   and torsion profiles, whereby the output force field of the first step (
   vdw-v1/result/optimize/force-field.offxml) was used as the starting point.

   The ids of the exact QC records (available from QCArchive) used in the fit can be found in the
   JSON serialized vdw-v1-td-opt-v3/optimization.json file.

   The full inputs and relevant outputs are located in the vdw-v1-td-opt-v3 directory.

Instructions for running each optimization can be found in the individual optimization directories.

If you are planning to re-run the optimizations from the provided inputs it is recommended that you create a
new conda environment using the recommended environment file:

conda env create --name openff-force-fields --file recommended-env.yaml

although the exact conda environment used to produce the fit can be found in each optimization directory
(conda-env.yaml).

See the openff-sage repository on GitHub for more information
and the scripts originally used to generate the inputs.