Skip to content

Latest commit

 

History

History
109 lines (93 loc) · 6.38 KB

Changelog.md

File metadata and controls

109 lines (93 loc) · 6.38 KB
Raw

Changelog

Until we get to a release version you may encounter different broken interface problem each time we increase a minor version.

Changes since v0.1.0rc0

  • SOAPify.HDF5er.isTrajectoryGroup now returns False with Datasets
  • Added some simple cli interfaces
  • Added examples for LENS and tSOAP and for generating hdf5 files
  • Added getSOAPSettings() for getting the SOAP organization from a hdf5 dataset
  • Added getTimeSOAPSimple() to give the user a shortcut to solve memory problem with large SOAP datasets
  • Now createUniverseFromSlice() returns also types
  • added precision selection for storing the trajectories and the SOAP fingerprints

Changes since v0.0.6

  • The v0.0.6 has been set flagged as the 'stable version'
  • Forced float64 precision in most of the calculations ← THIS IS A MAJOR CHANGE AND COULD IMPACT YOUR RESULTS
  • Changed the package name to cpctools, for storing it on https://pypi.org/
  • Removed the old and unused procedure for calculating the references
  • Compacted the workflows for a PR
  • Added the coveralls.io badge
  • Now documentation in rtd uses the standard rtd workflow
  • Refactored of the transition submodules, now support the 'window' and 'stride' concepts, added an entry to the documentation to explain it
  • Now saponifyMultipleTrajectories calls saponifyTrajectory
  • Moved HDF5er into SOAPify, to simplify the deployment
    • change your import HDF5er to import SOAPify.HDF5er as HDF5er in your old codes
    • HDF5er.HDF5erUtils → SOAPify.HDF5er.HDF5erUtils
    • HDF5er.HDF5To → SOAPify.HDF5er.HDF5To
    • HDF5er.ToHDF5 → SOAPify.HDF5er.ToHDF5
  • renamed some submodules to clarify the usage:
    • SOAPify.SOAPbase → SOAPify.distances
    • SOAPify.SOAPengine → SOAPify.engine
    • SOAPify.SOAPclassify → SOAPify.classify
    • SOAPify.SOAPTransitions → SOAPify.transitions
    • SOAPify.Saponify → SOAPify.saponify
  • renamed some functions to avoid naming collision, and to be more clear:
    • SOAPify.SOAPclassify.classify() → SOAPify.classify.applyClassification()
    • SOAPify.Saponify.saponifyGroup() → SOAPify.saponify.saponifyMultipleTrajectories()
    • SOAPify.Saponify.saponify() → SOAPify.saponify.saponifyTrajectory()
    • SOAPify.SOAPTransitions.normalizeMatrix() → SOAPify.transitions.normalizeMatrixByRow()
    • SOAPify.SOAPbase.KernelSoap() → SOAPify.distances.kernelSoap()
    • SOAPify.RemoveAtomIdentityFromEventTracker() → SOAPify.transitions.saponifyTrajectory()
  • Cleaned the code with pylint
  • Added LENS and tempoSOAP base functions

Changes since v0.0.4

  • Increased test coverage
  • Now if the box changes in a mda.Universe the ToHDF5 method saves it correctly
  • set quippy-ase compatibility to 0.9.10
  • Now tests prevents error in merging references
  • Checked and added python 3.10 compatibility

Changes since v0.0.3

  • getXYZfromTrajGroup now accept IO objects as inputs
  • WARNING: broken interface for getXYZfromTrajGroup, now it needs 3 inputs and the first is a file-like object
  • saveXYZfromTrajGroup and getXYZfromTrajGroup now can export comments per frames
  • transitionMatrixFromSOAPClassification now creates matrix with shape (n,n) and no more (n+1,n+1), where n is the lenght of the legend. The user will now need to address the errors in the legend, if needed
  • added calculateResidenceTimesFromClassification for calculating the residence times of the states during the MD
  • added trackStates for calculating the history of the evolution of the states in the MD
  • the result of trackStates can be used for calculating the residence times and the transition matrices
  • Now when appliyng soap, the created dataset will be given attributes that describe the parameters used for its creation
  • Removed some default values in function from Saponify and fillSOAPVectorFromdscribe
  • fillSOAPVectorFromdscribe now can fill soap vectors from multispecies calculations
  • changed slightly saponifyGroup and saponify: now they accept dscribe SOAP options as a dictionary, but not the sparse option
  • Now HDFTo.getXYZfromTrajGroup accepts slices as an option to export the trajectory
  • WARNING: broken interface for saponify
  • isTrajectoryGroup added to HDF5er to check if a group contain a trajectory in our format
  • now it is possible to select 'quippy' for calculating SOAP
  • tests reahul
  • fillSOAPVectorFromdscribe now is between 10 to 20 times faster than before
  • now name of the used engine is stored in the hdf5 file's SOAP dataset
  • added a small utility to make the nmax/lmax/rcut getter work regardless the compatible dscribe engine
  • now installation procedure ignores dscribe/quippy: you should install them separately, but you do not need to wait for their compilation if you only need the analysis features
  • created getXYZfromMDA for exporting exyz files (with personalized header and columns) from mda universe
  • removed a bug in getXYZfromHDF5: now boxes are exported correctly
  • some QoL improvements behind the curtains and in the tests
  • added CI for documentation
  • merged src/HDF5er/MDA2HDF5.py and src/HDF5er/ase2HDF5.py into src/HDF5er/ToHDF5.py
  • Heavy changes in the documentation
  • Now the documentation coverage should work
  • Documentation now should also be hosted by readthedocs.io

Changes since v0.0.2a

  • Only for Monospecies systems: added a small utility (fillSOAPVectorFromdscribe) that returns the complete SOAP vector from the simplified one from dscribe
  • Added a utility for normalize SOAP vectors
  • Added createReferencesFromTrajectory that creates a variables that stores SOAP references
  • set up a way to classify with soap with a different method thant the original idea
  • the new references now can be loaded/unloaded on an hdf5 file
  • added a patch for hdf5 imported files: workaround for mda not loading correctly non orthogonal boxes from lammps dumps

Changes since v0.0.2

  • Added the possiblity to export xyz files from the hdf5 trajectories, also with extra data columns
  • Improved performance for getXYZfromTrajGroup

Changes since v0.0.1

  • Tests
  • Changed default imports for HDF5er
  • Adding override option to HDF5er.MDA2HDF5
  • Added 3rd neighbours in the References
  • Added attributes to HDF5er.MDA2HDF5
  • In the referenceMaker: added the possibility to choose lmax and nmax for SOAP
  • Added the possibility to export to hdf5 slices of the trajectories