v0.1.1
What's Changed
Breaking Changes
- Minimum PyTorch raised from 2.4 to 2.8
- Minimum nvalchemi-toolkit-ops raised from 0.2 to 0.3
- Creating new TorchScript modules via
torch.jit.script()is no longer supported (legacy.jptloading preserved)
Changed
- Modernized nvalchemiops imports to v0.3 API (
nvalchemiops.torch.neighbors,nvalchemiops.torch.interactions.dispersion) - Replaced deprecated
torch.inverse()withtorch.linalg.inv() - Replaced
.transpose(-1, -2)with.mTfor matrix transpose - Conditional
torch.jit.optimized_executionfor ScriptModule only - Relaxed ASE dependency from
==3.27.0to>=3.27.0,<4
Fixed
- AIMNet2-Pd DFT reference corrected from wB97M-D3/CPCM to B97-3c/CPCM (THF)
- Model loading uses
weights_only=Trueby default (security) - Model download validates HTTP response before writing to disk
Documentation
- Modernized README with pip/uv/conda install instructions
- Updated TorchScript compatibility notes
Full Changelog: v0.1.0...v0.1.1