kinisi-1.1.1

What's Changed

• Move np.product to np.prod by @arm61 in #70
• Typo in the documentation by @arm61 in #71
• Updated mcmc docs by @Harry-Rich in #73
• Update vasp_d.ipynb docs by @Harry-Rich in #79
• Raise a warning and fix to documentation with respect to units by @arm61 in #83

Full Changelog: 1.1.0...1.1.1

kinisi-1.1.0

What's Changed

• Replace the _get_matrix function. by @jd15489 in #62
• Change to using matrix conditioning by @arm61 in #60
• Support for NPT unwrapping by @jd15489 in #64
• Add MSD test to aid future development by @jd15489 in #67
• Update timestep to time interval where appropriate by @arm61 in #68

Full Changelog: 1.0.0...1.1.0

kinisi-1.0.1

This is a bug-fix release. The bug is related to triclinic cells and produced an underestimate of the mean-squared displacement and, therefore, the diffusion coefficient.

Full Changelog: 1.0.0...1.0.1

kinisi-1.0.0

What's Changed

• Int now valid for get_indices by @arm61 in #38
• Move to cov_nearest by @arm61 in #48
• Add support for variance estimation using the block averaging approach. by @arm61 in #49
• Unwrapping FAQ by @arm61 in #50
• Change abbreviations in paper.md by @zhubonan in #51
• Correct typos in paper.md by @jd15489 in #53
• Faster implementation of the get_disp function by @jd15489 in #54
• Adds centre of geometry and centre of mass functionality by @arm61 in #47
• Cog remove history by @arm61 in #56
• Fix off-by-one error for framework indices by @jd15489 in #59

New Contributors

• @zhubonan made their first contribution in #51
• @jd15489 made their first contribution in #53

Full Changelog: 0.7.2...1.0.0

kinisi-0.7.2

Changes

• @arm61 Revert the posterior predictive use for the extrapolation, as it is not possible given there is no known uncertainty in the extrapolated value.

kinisi-0.7.1

Changes

• @arm61 Support for the posterior predictive of the MSD distribution.
• @arm61 Start using uravu version 1.3.0 which supports the posterior predictive of the Relationship object. This is used to extrapolate the diffusion at a given temperature for the StandardArrhenius and SuperArrhenius objects.

kinisi-0.7.0

Changes

• @arm61 - An argument has been added to the diffusion method. It is now necessary to pass the first time interval (in kinisi units of picoseconds) that should be used in the diffusion analysis. This should be greater than the start of the diffusive regime.
• @alexsquires - Support for ASE objects has been added.
• @alexsquires - Improvements to the saving and loading .hdf file functionality.
• @arm61 - Documentation improvements.

kinisi-0.6.5

Changes

• Bug fixes

kinisi-0.6.4

Changes

• Bug fixes

kinisi-0.6.3

Changes

• The scaled variance is now used to estimate the variance of the mean-squared displacement. Bootstrapping can still be used to find the distribution of the MSD with the bootstrap=True flag