Releases: nlesc-nano/auto-FOX
Releases · nlesc-nano/auto-FOX
Auto-FOX 1.0.0b1
What's Changed
- DEP: Bump qmflows and deal with deprecations by @BvB93 in #280
- ENH: Issue a warning in the psf recipe if the qd or ligands don't have any bonds by @BvB93 in #281
- BUG: Fix an issue with parsing multiple ligands in
generate_psf2
by @BvB93 in #282 - MAINT: General maintenance in preperation for Auto-FOX 1.0.0 by @BvB93 in #283
Full Changelog: 0.10.2...1.0.0b1
Auto-FOX 0.10.2
What's Changed
- REL: Update
master
branch after release (0.10.1) by @BvB93 in #261 - BUG: Fix atom-pairs not being sorted if certain parameters are guessed by @BvB93 in #262
- ENH: Add a workflow for computing Debye scattering by @BvB93 in #263
- BUG: Fix PES-averaged ARMC failing to create multiple
ParamMapping
columns by @BvB93 in #264 - BUG: Fix PA-ARMC not correctly reshaping the acceptance array by @BvB93 in #265
- BUG: Fix the param columns being incorrectly written to the .hdf5 file by @BvB93 in #266
- BUG: Transpose the axes when returning in
get_multi_lig_center
by @BvB93 in #267 - ENH: Store the git commit hash of auto-FOX in the ARMC .hdf5 file by @BvB93 in #268
- ENH: Print the offending atoms whenever the ARMC constraint-resolver fails by @BvB93 in #269
- BUG: Fix the ARMC constraints solver falsely rejecting moves due to 0 division by @BvB93 in #270
- ENH: Export warnings to the logger when running ARMC by @BvB93 in #271
- BUG: Fix
DataFrame.loc
indexing failing when passed a 2-tuple by @BvB93 in #272 - BUG: Fix only settings not being properly updated for PES-averaged ARMC by @BvB93 in #273
- TST: Enable windows tests by @BvB93 in #274
- BUG: Fix improper dihedral-indices being incorrectly sorted by @BvB93 in #275
- ENH: Allow the .xyz parser to handle empty lines by @BvB93 in #277
- ENH: Add special casing for zero-division in the ARMC error func by @BvB93 in #278
- BUG: Do not directly rely on PLAMS for obtaining the Boost
ArgumentError
type by @BvB93 in #279
Full Changelog: 0.10.1...0.10.2
0.10.1
Auto-FOX 0.10.0
What's Changed
- ENH: Add a new
MultiMolecule
method for constructing supercells by @BvB93 in #242 - TST: Refactor the
get_non_bonded
tests by @BvB93 in #243 - BUG: Fixed an issue wherein init_power_spectrum could raise when the atom subset is specified by @BvB93 in #245
- BLD: Bump the PLAMS minimum version to >= 1.5.1 by @BvB93 in #247
- BUG: Fixed an issue wherein net charges weren't properly conserved in PES-averaged ARMC by @BvB93 in #235
- ENH: Allow users to pass custom error functions by @BvB93 in #249
- ENH: Add a
max
-based error function by @BvB93 in #250 - BLD: Make
rdkit
an optional dependency (again) by @BvB93 in #251 - ENH: Add two new error functions by @BvB93 in #252
- MAINT,ENH: Rework the auto-FOX properties system by @BvB93 in #253
- BUG: Fixed an issue wherein auto-FOX would not check the status of jobs by @BvB93 in #254
- BUG: Fixed an issue wherein dataframes were summed along the wrong axis by @BvB93 in #255
- ENH: Add error functions based on non-mse errors by @BvB93 in #256
- ENH: Add the new
atom_pairs
keyword toinit_rdf
andinit_adf
by @BvB93 in #257 - ENH: Round the net charge to the nearest integer by @BvB93 in #258
- REL: Release Auto-FOX 0.10.0 by @BvB93 in #259
Full Changelog: 0.9.1...0.10.0
Auto-FOX 0.9.1
- Added the new
segment_dict
parameter to thePSFContainer.generate_x()
functions. - Added the new
PSFContainer.sort_values
method. - Store metadata and net charges for each individual system in PES-averaged ARMC and ARMCPT.
- Fixed an issue wherein frozen atomic charges were ignored when not explicitly specified.
- Fixed in issue wherein PSFContainer.sort_values did not not update the residue ID.
- Fixed an issue wherein guessed parameters could overwrite ones that were explicitly specified.
- Fixed an issue wherein frozen parameters weren't properly sorted.
- Fixed an issue wherein the mean pair-density was computed using lattice vectors in non-periodic calculations.
Auto-FOX 0.9.0
- Fixed an issue wherein
PSFContainer.to_atom_alias_dict
would assign incorrect indices to atom aliases. - Fixed an issue wherein treating an entire ARMC parameter block as frozen could raise.
- Added the
FOX.MultiMolecule.lattice
property for storing the lattice vectors of periodic systems. - Added
FOX.io.lattice_from_cell
, a function for reading lattice vectors from CP2K .cell files. - Added support for periodic ADF and RDF calculations.
- Added the new
job.lattice
keyword to the ARMC workflow for specifying lattice vectors of the reference system. - Lattice vectors are now preserved when interconverting between PLAMS, ASE and Auto-FOX.
- Read the pressure from the CP2K .out file when calculating the bulk modulus, rather than calculating it from scratch.
Auto-FOX 0.9
Auto-FOX 0.9.1
- Added the new
segment_dict
parameter to thePSFContainer.generate_x()
functions. - Added the new
PSFContainer.sort_values
method. - Store metadata and net charges for each individual system in PES-averaged ARMC and ARMCPT.
- Fixed an issue wherein frozen atomic charges were ignored when not explicitly specified.
- Fixed in issue wherein PSFContainer.sort_values did not not update the residue ID.
- Fixed an issue wherein guessed parameters could overwrite ones that were explicitly specified.
- Fixed an issue wherein frozen parameters weren't properly sorted.
- Fixed an issue wherein the mean pair-density was computed using lattice vectors in non-periodic calculations.
Auto-FOX 0.9.0
- Fixed an issue wherein
PSFContainer.to_atom_alias_dict
would assign incorrect indices to atom aliases. - Fixed an issue wherein treating an entire ARMC parameter block as frozen could raise.
- Added the
FOX.MultiMolecule.lattice
property for storing the lattice vectors of periodic systems. - Added
FOX.io.lattice_from_cell
, a function for reading lattice vectors from CP2K .cell files. - Added support for periodic ADF and RDF calculations.
- Added the new
job.lattice
keyword to the ARMC workflow for specifying lattice vectors of the reference system. - Lattice vectors are now preserved when interconverting between PLAMS, ASE and Auto-FOX.
- Read the pressure from the CP2K .out file when calculating the bulk modulus, rather than calculating it from scratch.
Auto-FOX 0.8.12
- Added a recipe for calculating the similarity between 2 MD trajectories.
- Added improved support for custom atom types.
- Made the ligand optional in the
FOX.recipes.generate_psf()
recipe. - Fixed an issue where non-absolute distances were used when calculating the pressure.
- Fixed an issue where non-charge parameters were updated incorrectly.
- Fixed an issue where parameter guessing could fail if no
.prm
file was provided.
Auto-FOX 0.8.11
- Fixed an issue where the .hdf5 status would not be properly cleared (if necessary).
- Grab the cell parameters from previous jobs if applicable.
- More thoroughly check the
qmflows.Result
status before accessing it. - Automatically determine appropriate chunk sizes when calculating the RDF.
- Added a new sub-module dedicated to the calculation of properties:
FOX.properties
. - Fixed an issue where parsing the
unit
would fail when parameter guessing. - Fixed an issue where writing
nan
toarmc.xyz.hdf5
would fail.
Auto-FOX 0.8.10
- Reorganized the dataframes in
FOX.armc.ParamMapping
. - Added the
pes_validation
keyword. Used for constructing a set of PES descriptors for the purpose of validation. - Dissallow the specification of duplicate .yaml keys.
- Allow users to provide unique settings for each molecule in PES-averaged ARMC.
- Added tests using CP2K, python 3.9, pre-releases and the minimum supported version of dependencies.
- Removed the plotting and .csv-related entry points.
- Added the
param.validation.enforce_constraints
keyword; used for checking whether or not initially supplied parameters satisfy all user-provided constraints. - Added the
unit
field to theparam_metadata
dataset. - Added the new
pes.block.ref
keyword; used for constructing PES descriptors fromqmflows.Result
objects rather than theFOX.MultiMolecule
instances constructed therefrom.