• Fix bug of analytical stress of PBC (#221)
• Add support for analytical hessian and vibrational analysis (#222)

• Fix a bug on PBC (#214)
• Add support for analytical stress (#218)

• Fix ase interface on GPU (#212)

• Further performance improvements (#202, #204, #205, #206,#207)
• Use a stable API torch.unique_consecutive to replace the temporary unique in most recent nightly PyTorch (#209)

• Rewrite AEVComputer to improve performance and reduce memory usage (#197)

Note that this release use a very new PyTorch operator torch.repeat_interleave and a very new temporary API torch._unique_dim2_temporary_will_remove_soon. Please make sure you update your nightly PyTorch before updating TorchANI. New version of TorchANI will be released when torch._unique_dim2_temporary_will_remove_soon is removed to make sure TorchANI always work with latest nightly PyTorch.

• Inference on Python 2.7 (#171 , #172)

• Added ANI-1ccx model. (#167)
• Added model zoo API. (#169)

• Fix a bug of ASE interface giving zero forces when box size is 0 (#165)
• Some fixes on Windows (#161)
• Fix docs about installation (#163)

Basically same as 0.2
Use bdist to deploy to PYPI to reduce package size

• ASE interface
• YAML constructor for NeuroChem Trainer
• Version tag torchani.__version__