metatomic-torch v0.1.8

This is a new release of metatomic-torch, with the following changes:

Added

• Added the ability for models to request extra inputs from the simulation engine, on top of positions/cell/types. These extra inputs are requested by the model and stored on the system by the engine (like for neighbor lists). They are provided as TensorMap, stored in the system's data, and follow the same metadata structure as the standard outputs (anything that can be an output can also be an input).
• Added support for charges, masses, velocities as standard model inputs and outputs
• Added support for PyTorch v2.10

Changed

• Properties in standard output/inputs should now consistently use singular (i.e. position instead of positions). The plural names are deprecated and will be remove in the future.

metatomic-torch v0.1.7

This is a small patch release for metatomic-torch, with the following changes:

• Fixed metatomic.torch.ase_calculator.MetatomicCalculator to work even when a model does not have an energy output (#121)
• Added support for loading TorchScript extensions from deepmd-kit (#98)

metatomic-torch v0.1.6

Added

• Added support for torch v2.9
• ModelOutput now has a description field, to carry more information about a given output.
• Added the pick_output function that can be used by simulation engines to pick the correct output based on what's available inside a model and which variant (if any) the user requested.
• Added metatomic.torch.ase_calculator.SymmetrizedCalculator, an ASE calculator that can average energy calculations over the O(3), SO(3) or space group of a crystal, to make unconstrained, non-equivariant models rotationally equivariant up to the integration order.

Changed

• The pick_device function now returns a torch::DeviceType instead of a string in C++. This does not affect the Python API.
• It is now possible to construct an AtomisticModel with a torch module already compiled with TorchScript.

Fixed

• Variants can now be used with non-standard outputs as well (#105)

Removed

• We dropped support for Python 3.9, and now requires at least Python 3.10

metatomic-torch v0.1.5

This is a patch release of metatomic-torch, containing the following changes:

Added

• Added support for variants of outputs, allowing one model to define all of energy, energy/pbe0 and energy/lj_6_12 as outputs. Which output to use is then picked at runtime by the simulation engine.
• Two new functions, metatomic.torch.save_buffer and metatomic.torch.load_system_buffer, allow to serialize and deserialize System objects to and from torch.Tensor

Changed

• metatomic.torch.save and metatomic.torch.load_system are now implemented in C++
• We now requires at least cmake v3.22 to compile metatomic-torch

metatomic-torch v0.1.4

A new release of metatomic-torch adding support for the following:

• The ASE MetatomicCalculator will now send warnings if the model predicts a high per-atom uncertainty for its energy output.
• We now have two new standard outputs: positions and momenta, which can be used for direct structure prediction and bypassing time integration.
• System.to accepts a non_blocking argument, with the same meaning as torch.Tensor.to.

---

• The code is now compatible with metatensor-torch v0.8.0
• The code is now compatible torch v2.8

metatomic-torch v0.1.3

This is a patch release for metatomic-torch, with the following changes:

• The logic to detect OpenMP dependencies in TorchScript extensions now takes into account the user's site-package directory (#65)
• metatomic.torch.ase_calculator is now lazy-loaded, and can be accessed directly after importing metatomic.torch (#59)

metatomic-torch v0.1.2

This is a patch release of metatomic-torch, fixing the following issues:

• register_autograd_neighbors is now kept in the code by the TorchScript compiler. It was previously silently removed.
• When running ase_calculator.Metatomic with non_conservative=True, we no longer crash for NPT simulations.

metatomic-torch v0.1.1

This is a patch update to metatomic-torch, with the following change:

• metatomic_torch can now be built as part of the same cmake project as metatensor and metatensor_torch (#33)

metatomic v0.1.0

metatomic-v0.1.0

Release metatomic and metatomic-torch v0.1.0

metatomic-torch v0.1.0

This is the first release of metatomic-torch as an independent package. This code was previously part of metatensor-torch, in the atomistic sub-module.