v2025.8.1

Changed

• Checkpoints for fine-tuning files are now passed from the options.yaml.

v2025.7

Changed

• Metatrain is now built on top of metatomic instead of
  metatensor.torch.atomistic. Please refer to https://docs.metatensor.org/metatomic/
  to find how to use the new models.

v2025.6

Fixed

• PET can now evaluate on single-atom structures without crashing
• The metatrain dataloader doesn't load all batches ahead of each epoch anymore

Added

• NanoPET and PET can now train on non-conservative stresses
• Users can now choose the name of the extension directory in mtt train and
  mtt export via the --extensions (or -e) option
• Update to metatensor-torch-0.7.6, adding support for torch 2.7
• PET now supports gradient clipping as a new training hyperparameter

Changed

• Training and exporting models without extensions will no longer lead to the creation
  of an empty directory for the extensions
• The SOAP-BPNN model now uses torch-spex instead of featomic as its SOAP
  backend
• PET from the previous version is now deprecated and accessible as
  deprecated.pet, while the old NativePET (experimental.nativepet) is
  now called PET (pet from training option files)
• The Angstrom character is now represented as A and not Å in the training logs

v2025.5

This is the latest metatrain release 🚀 including the following changes.

Fixed

• Fix more composition model issues

Added

• Update to metatensor-torch-0.7.5 to allow training on non_conservative_forces
  and non_conservative_stress targets
• Add NativePET as a readable, efficient, backward-compatible PET implementation
• Added Wandb logger
• Save loss history in a .csv file

v2025.4

Changed

• upgraded to metatensor.torch 0.7.4, which gives access to batched ASE evaluation

v2025.3

Fixed

• Fixed a bug in the composition model, affecting SOAP-BPNN and nanoPET

Changed

• metatrain.util.io.load_model does not copy a remote model to the current
  directory.

v2025.2

This is the latest metatrain release 🚀 including the following changes.

Added

• Implement a long-range featurizer as a utility for all models
• Speed up system preparation

Changed

• Remove biases in SOAP-BPNN's linear layers

Fixed

• Fix NanoPET multi-GPU error message
• Fix device for fixed composition weights

Full Changelog: v2025.1...v2025.2

v2025.1

This is the latest metatrain release 🚀 including the following changes.

Added

• Support for Python 3.13 and ase >= 3.23 @PicoCentauri in #497

Fixed

• Some irrelevant autograd warnings @frostedoyster in #494

Full Changelog: v2025.0...v2025.1

v2025.0

This is the first release of metatrain 🚀🎉