v0.13.0

[0.13.0]

Added

• reEWC fine-tuning with forgetting prevention for single-modal models: optional experience replay (rehearsal, load_memory_path, mem_batch_size, mem_ratio) and an Elastic Weight Consolidation penalty from a precomputed Fisher matrix (continue.fisher_information, continue.opt_params, continue.ewc_lambda), plus a cosineannealingwarmuplr scheduler
• Support OpenEquivariance
• Per-atom stress (atomic virial) support in LAMMPS pair_e3gnn and ASE calculator
• compute_atomic_virial option in SevenNetCalculator
• Add Batch D3 and SevenNetD3Model in Torch-Sim interface

Changed

• [Breaking] Rename optional dependency group mliap into mliap12 (reflecting its CUDA 12.x dependency).
• Add cueq13 and mliap13 optional dependency groups for CUDA 13.x.
• [Breaking] torchscript file_type is no longer supported in SevenNetCalculator.
• LAMMPS pair_e3gnn refactored to use pair-wise force (dE/dr) instead of position-based gradient.
• Deploy no longer replaces force_output with ForceStressOutput; force/stress computed in LAMMPS C++ side.
• [Breaking] Accelerator None options (use the same setting as checkpoint) are removed and default is changed to False.

Fixed

• Omni (flash TP trained checkpoint) loads w/o an error for flashTP disabled environment.
• Load FlashTP-saved checkpoints (e.g. SevenNet-Nano) when FlashTP is unavailable by falling back to the e3nn backend, so they work for inference and fine-tuning without FlashTP installed. An explicit enable_flash=True still fails loud.

New Contributors

• @gshs12051 made their first contribution in #297
• @kskjs1203 made their first contribution in #312

Full Changelog: v.0.12.1...v0.13.0