Notes
This is the first release of SchNetPack 2.0 which uses the Hydra configuration framework, Pytorch Lightning and a new indexing scheme.
It also includes an improved data pipeline, modules for equivariant neural networks and a PyTorch implementation of molecular dynamics.
What's Changed
* Inital commit for v1 rewrite by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/267
* center transformations by Stefaanhess in https://github.com/atomistic-machine-learning/schnetpack/pull/271
* Mg/torch env by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/274
* Kts/qm9datamodule by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/275
* Training script driven by Hydra+Lightning by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/277
* Implement training of potential energy surfaces by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/278
* Add split file by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/282
* PaiNN representation by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/284
* Postprocessors and TorchScript by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/285
* Mg/symfuncs by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/287
* Initial API docs by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/293
* Fix problem with transforms in new lightning version by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/294
* stress and custom experiment by Stefaanhess in https://github.com/atomistic-machine-learning/schnetpack/pull/292
* Update docs (and fix postprocess bug) by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/295
* Mg/calculators by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/296
* Refactor models by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/297
* fix small bug by Stefaanhess in https://github.com/atomistic-machine-learning/schnetpack/pull/298
* Sh/datasets by Stefaanhess in https://github.com/atomistic-machine-learning/schnetpack/pull/300
* Proposal for MultiPropertyModel by Stefaanhess in https://github.com/atomistic-machine-learning/schnetpack/pull/299
* add epsilon to painn by Stefaanhess in https://github.com/atomistic-machine-learning/schnetpack/pull/301
* Some minor updates by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/305
* Fix API docs by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/306
* Unify model classes & refactor configs by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/309
* Fix ModelOutput package by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/310
* add bessel representation for future painn usage by Divide-By-0 in https://github.com/atomistic-machine-learning/schnetpack/pull/311
* Update examples & add data workdir by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/314
* Dipole moment & polarizability by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/316
* Dynamics caching neighborlist by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/320
* Dev by jnsLs in https://github.com/atomistic-machine-learning/schnetpack/pull/322
* Add long range cutoff by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/325
* Mg/md by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/315
* Refactor configs and add predict script by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/334
* Fix install bug by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/335
* Add automatic position derivatives by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/341
* Update QM9 tutorial by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/347
* ZBL Potential, Electrostatics (+Ewald summation) and stress tensor fixes by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/349
* Fixes to some MD desfaults and torchscript issues by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/350
* Disable automatic use of torchscript in MD calculators by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/351
* Response properties and field representations by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/339
* Fixed derivative graph settings for basic response properties by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/352
* Fixed sign for shift type cutoff by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/353
* Refactor AtomisticModel by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/354
* Fix bug when using ddp with set run.id by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/359
* Fix mixing bias by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/365
* Fix mixing residual by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/366
* fixed aggregation_mode bug for avg pooling by jnsLs in https://github.com/atomistic-machine-learning/schnetpack/pull/361
* Sh/ep device by Stefaanhess in https://github.com/atomistic-machine-learning/schnetpack/pull/360
* Fix AtomsDataSubset for use inside ConcatAtomsData by chgaul in https://github.com/atomistic-machine-learning/schnetpack/pull/357
* Add learning rate warmup & SGD config by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/370
* Fixed creation of subset in BaseAtomsData by NiklasGebauer in https://github.com/atomistic-machine-learning/schnetpack/pull/369
* Fix DDP training by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/372
* Updated for new yaml behavior by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/375
* Nwag/comment-update by NiklasGebauer in https://github.com/atomistic-machine-learning/schnetpack/pull/374
* Fix OMDB _convert dataset preparation by bartolsthoorn in https://github.com/atomistic-machine-learning/schnetpack/pull/373
* Fix pin_memory and some deprecation warnings by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/378
* Update tutorials and filter outputs by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/377
* Add conversion script for old datasets by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/380
* Fixed bug in map_properties by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/381
* Fix testing bc lightning API changed by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/382
* Updated MD docstrings and tutorial by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/383
* Wrapping of atom positions under PBC by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/385
* Fixed energy logging for multiple molecules by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/386
* Some refactoring and cleanup by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/387
* Retrained ethanol model for new postprocessing convention by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/388
* Added `on_step_finalize` in MD simulation hooks by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/390
* Updated weight init in FieldSchNet representation by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/391
* Added tmpdir functionality for MDs, fixed calculator bug by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/395
* Improved config loading for MD by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/396
* Fixed criterion for recomputing MD neighborlists by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/398
* Fixed using subset in ASEAtomsData.iter_properties by NiklasGebauer in https://github.com/atomistic-machine-learning/schnetpack/pull/393
* update deprecated code to new torch version by Stefaanhess in https://github.com/atomistic-machine-learning/schnetpack/pull/399
* Fix strain input module by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/403
* Add resolver for tmp directory by NiklasGebauer in https://github.com/atomistic-machine-learning/schnetpack/pull/406
* Fix 401 by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/404
* Added tempfile import for custom tmpdir resolver by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/408
* Added routine to NeighborListMD to properly filter out pairs due to the buffer region by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/409
* Removed `n_out` argument from `DipoleMoment` and `Polarizability` layer docstrings by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/411
* Fix PyTorch Lightning deprecations by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/414
* Consider only a selection of atomic forces in training, validation and testing, and ASE neighborlist with skin implemented by jnsLs in https://github.com/atomistic-machine-learning/schnetpack/pull/405
* Added a few classes util for structure relaxations (in particular MOMONANO) by jnsLs in https://github.com/atomistic-machine-learning/schnetpack/pull/415
* New datasets and fixes by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/417
* Fixed inverted grad context for calculator by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/419
* Update to ASE interface and SchNetPack ASE calculator by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/418
* Added neighborlist from MatScipy package by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/421
* Upgrade to Hydra 1.2 by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/420
* Upgrade Hydra configs by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/424
* Fixed issue with merging loaded configs by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/423
* Fix charge correction in dipole moment by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/425
* Merge SchNetPack2 into master branch by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/426
* additional inputs are now an attribute of nbh list postprocessing cla… by jnsLs in https://github.com/atomistic-machine-learning/schnetpack/pull/427
* Fix error in AddOffsets by NiklasGebauer in https://github.com/atomistic-machine-learning/schnetpack/pull/432
* Fixed a problem with hyperparameter logging by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/429
* Remove some unneccesary code and simplify worker defaults by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/434
* Bump protobuf from 3.20.1 to 3.20.2 in /docs by dependabot in https://github.com/atomistic-machine-learning/schnetpack/pull/437
* Bump protobuf from 3.20.1 to 3.20.2 by dependabot in https://github.com/atomistic-machine-learning/schnetpack/pull/436
* Add infos for atomref by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/439
* implemented torch lbfgs optimizer that allows for relaxation of multiple structures in parallel by jnsLs in https://github.com/atomistic-machine-learning/schnetpack/pull/438
* Add workaround for pytorch issue with serializing dtype by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/442
* Make compatible with older PyTorch versions by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/443
* maxstep can be used to renormalize the integration steps by jnsLs in https://github.com/atomistic-machine-learning/schnetpack/pull/441
* Add config docs by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/444
* NeighborListWrapper is removed by jnsLs in https://github.com/atomistic-machine-learning/schnetpack/pull/446
* Add SO3 ops and layers by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/445
* Moved and adapted CLI docs for MD. by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/447
* SO3 representation by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/448
* Fix ScaleProperty transform by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/450
* Add mixing layers to SO3net by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/451
* Fix load_properties by NiklasGebauer in https://github.com/atomistic-machine-learning/schnetpack/pull/454
* Fix metrics computation by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/455
* Separate metric objects for train/val/test by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/456
* Updates for response experiments / NMR utilities by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/457
* Cached NBL takes nbh_transforms by jnsLs in https://github.com/atomistic-machine-learning/schnetpack/pull/458
* Fix configs and docstrings by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/459
* Fix EMA, upgrade to PL 1.8 by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/460
* Refactored MD17 DataModule and added MD22 datset by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/461
* Readd residual by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/463
* Make return of vector representation in SO3net optional by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/465
* Make transforms independently usable from datamodule by ktschuett in https://github.com/atomistic-machine-learning/schnetpack/pull/464
* fixes initialization bug in so3net by Stefaanhess in https://github.com/atomistic-machine-learning/schnetpack/pull/467
* added ASEBatchwiseLBFGS, updated tutorial by jnsLs in https://github.com/atomistic-machine-learning/schnetpack/pull/466
* updated tutorial 5 by Stefaanhess in https://github.com/atomistic-machine-learning/schnetpack/pull/468
* Fixed typos in tutorial 5 by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/469
* Added reference to schnetpack-gschnet to docs... by NiklasGebauer in https://github.com/atomistic-machine-learning/schnetpack/pull/470
* cleaned up batchwise optimizer by jnsLs in https://github.com/atomistic-machine-learning/schnetpack/pull/471
* updated requirements.txt by Stefaanhess in https://github.com/atomistic-machine-learning/schnetpack/pull/475
* added trained model to tutorial by Stefaanhess in https://github.com/atomistic-machine-learning/schnetpack/pull/474
* Updated docs and fixed problems by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/477
* Switched to matscipy pypi package by mgastegger in https://github.com/atomistic-machine-learning/schnetpack/pull/478
New Contributors
* Divide-By-0 made their first contribution in https://github.com/atomistic-machine-learning/schnetpack/pull/311
* jnsLs made their first contribution in https://github.com/atomistic-machine-learning/schnetpack/pull/322
* chgaul made their first contribution in https://github.com/atomistic-machine-learning/schnetpack/pull/357
* NiklasGebauer made their first contribution in https://github.com/atomistic-machine-learning/schnetpack/pull/369
* dependabot made their first contribution in https://github.com/atomistic-machine-learning/schnetpack/pull/437
**Full Changelog**: https://github.com/atomistic-machine-learning/schnetpack/compare/v1.0.0...v2.0.0