v2.0.6

• Bug fix for CHGnet loading.

v2.0.5

• Improved error messages for backend/model mismatch. Try to transparently handle simple situations.

v2.0.4

• Bug fix for matgl.graph.data and matgl.graph.converter imports for different backends.

v2.0.3

• Bug fix for matgl.ext.pymatgen import for different backends.

v2.0.2

• QET (Charge-Equilibrated TensorNet) architecture and pre-trained weights are added!
• Begun a migration to Pytorch-Geometric over the now-deprecated DGL. So far, only vanilla TensorNet has been
  implemented in PYG). DGL models still work but require a manual setup (change of backend and installation of DGL).

v2.0.1

• QET (Charge-Equilibrated TensorNet) architecture and pre-trained weights are added!
• Begun a migration to Pytorch-Geometric over the now-deprecated DGL. So far, only vanilla TensorNet has been
  implemented in PYG). DGL models still work but require a manual setup (change of backend and installation of DGL).

v2.0.0

• QET (Charge-Equilibrated TensorNet) architecture and pre-trained weights are added!
• Begun a migration to Pytorch-Geometric over the now-deprecated DGL. So far, only vanilla TensorNet has been
  implemented in PYG). DGL models still work but require a manual setup (change of backend and installation of DGL).

v1.3.0

This release includes pretrained molecular potentials and implements the preliminary PyG framework for future development.

What's Changed

• publish wheel as well as sdist by @dimbleby in #597
• Example notebook for fine-tuning M3GNet potential on a customized dataset with DIRECT sampling by @kenko911 in #599
• pre-commit autoupdate by @pre-commit-ci[bot] in #602
• First stage for PyG TensorNet implementation done by @kenko911 in #604
• Update lightning requirement from <=2.5.1 to <=2.5.1.post0 by @dependabot[bot] in #606
• Add element_types kwarg to example notebook by @Andrew-S-Rosen in #611
• Update lightning requirement from <=2.5.1.post0 to <=2.6.0.dev20250629 by @dependabot[bot] in #610
• Ruff fix in MGLDataset class by @kenko911 in #614
• Update lightning requirement from <=2.6.0.dev20250629 to <=2.6.0.dev20250706 by @dependabot[bot] in #615
• pre-commit autoupdate by @pre-commit-ci[bot] in #616
• TensorNet PyG added by @kenko911 in #617
• Update lightning requirement from <=2.6.0.dev20250706 to <=2.6.0.dev20250713 by @dependabot[bot] in #618
• Fix the bug in the united test for ase.py by @kenko911 in #621
• Update lightning requirement from <=2.6.0.dev20250713 to <=2.6.0.dev20250720 by @dependabot[bot] in #622
• Bump nokogiri from 1.18.8 to 1.18.9 in /docs by @dependabot[bot] in #623
• Update lightning requirement from <=2.6.0.dev20250720 to <=2.6.0.dev20250727 by @dependabot[bot] in #624
• pre-commit autoupdate by @pre-commit-ci[bot] in #627
• Update lightning requirement from <=2.6.0.dev20250727 to <=2.6.0.dev20250803 by @dependabot[bot] in #629
• Added a dropout argument to the M3GNET constructor by @miicck in #628
• Update lightning requirement from <=2.6.0.dev20250803 to <=2.6.0.dev20250810 by @dependabot[bot] in #631
• Pretrained molecular potentials from MatGL paper and additional thermostats for ASE MD added by @kenko911 in #633

New Contributors

• @dimbleby made their first contribution in #597
• @miicck made their first contribution in #628

Full Changelog: v1.2.7...v1.3.0

v1.2.7

• Use original custom RemoteFile rather than fsspec, which is very finicky with SSL connections.
• _create_directed_line_graph error handling (@bowen-bd)
• Update Import Alias for lightning (@jcwang587)
• Add nvt_nose_hoover to MD ensemble (@bowen-bd)
• Allow training of magmom when no line graph presents (@bowen-bd)
• Allow disable BondGraph in CHGNet (@bowen-bd)

v1.2.6

• Fix missing torchdata dependency for Linux.