NNPOps 0.6

What's Changed

• Add Self-Hosted AWS GPU Runner by @mikemhenry in #100
• Particle Mesh Ewald by @peastman in #103
• Fix CMake not adding some tests by @RaulPPelaez in #104
• Make some tests use deterministic torch algorithms by @RaulPPelaez in #108

New Contributors

• @mikemhenry made their first contribution in #100

Full Changelog: v0.5...v0.6

NNPOps 0.5

What's Changed

• Update CI versions by @raimis in #88
• Additions to .gitignore by @RaulPPelaez in #90
• Fix neighbourlist gradient calculation on CPU by @sef43 in #91
• Move all torch.ops.load calls to the __init__.py scripts by @RaulPPelaez in #89
• Fix incorrect decorator in NNPOps by @RaulPPelaez in #93
• Fix CMake not updating tests correctly after first run by @RaulPPelaez in #97
• Add error checking to CUDA version of getNeighborPairs by @RaulPPelaez in #80
• Support for pytorch 2.0 by @RaulPPelaez in #94

Full Changelog: v0.4...v0.5

NNPOps 0.4

This release fixes a bug in the periodic boundary conditions in NNPOps.SymmetryFunctions.TorchANISymmetryFunctions and improves the build system.

What's Changed

• Improve the build system to use find_package(Python) instead of find_package(PythonLibs) by @RaulPPelaez in #81
• Fix a bug in the periodic boundary conditions in ANI SymmetryFunctions by @sef43 in #83

New Contributors

• @RaulPPelaez made their first contribution in #81
• @sef43 made their first contribution in #83

Full Changelog: v0.3...v0.4

NNPOps 0.3

This release brings the nearest neighbor operation NNPOps.neighbors.getNeighborPairs and a few bug fixes.

What's Changed

• Update README by @raimis in #52
• Fix and improve CI by @raimis in #56
• Nearest neighbor operation by @raimis in #58
• getNeighborPairs compute the displacement vector by @raimis in #61
• Use Mamba for CI by @raimis in #64
• Refactor the common code by @raimis in #65
• Fix the selection of the activation function in CFConv by @raimis in #67
• Workaround for pytorch bug by @peastman in #71
• Align SpeciesConverter return type with TorchANI version by @scal444 in #74
• Update the CI versions by @raimis in #75
• getNeighborPairs() supports periodic boundary conditions by @peastman in #70

New Contributors

• @scal444 made their first contribution in #74

Full Changelog: v0.2...v0.3

NNPOps 0.2

This release brings a full support of TorchANI via NNPOps.OptimizeTorchANI. Additionally, the PyTorch wrappers for the continuous-filter convolution (CFConv) operations have been implemented, which will be used to accelerate TorchMD-Net.

What's Changed

• Optimization to SchNet by @peastman in #42
• Fix the execution of ANISymmetryFunctions on a non-default CUDA stream by @raimis in #37
• Move the TorchANI imports from the top level by @raimis in #45
• Optimize EnergyShifter of TorchANI by @raimis in #38
• Optimize SpeciesConverter of TorchANI by @raimis in #39
• Optimize TorchANI by @raimis in #41
• Enable CPU-only builds by @raimis in #48
• Improve the wrapper of ANISymmetryFunctions by @raimis in #47
• PyTorch wrapper for SchNet operations by @raimis in #40
• Refactor the CUDA error check by @raimis in #49

Full Changelog: v0.1...v0.2

NNPOps 0.1

What's Changed

• Update to TorchANI 2.2.2 by @raimis in #35

Full Changelog: v0.0...v0.1

NNPOps 0.0

The first official release! It provides a minimal functionality to accelerate the ANI models from TorchANI.

What's Changed

• CUDA implementation of ANISymmetryFunctions by @peastman in #4
• Optimizations to CudaANISymmetryFunctions by @peastman in #7
• Further optimizations to CudaANISymmetryFunctions by @peastman in #8
• Fix a memory leak in CudaANISymmetryFunctions by @raimis in #9
• Added README files and benchmark code by @peastman in #12
• PyTorch wrapper by @raimis in #5
• mdtraj xyz unit by @yueyericardo in #17
• CFConv layer for SchNet by @peastman in #18
• Added option to use tanh activation by @peastman in #23
• A better PyTorch wrapper by @raimis in #19
• Reorganize repo by @raimis in #33
• Batched NNs for TorchANI by @raimis in #13
• CI for NNPOps by @raimis in #34

New Contributors

• @peastman made their first contribution in #4
• @raimis made their first contribution in #9
• @yueyericardo made their first contribution in #17