ase merge requestshttps://staging.gitlab.com/ase/ase/-/merge_requests2016-09-15T15:06:54Zhttps://staging.gitlab.com/ase/ase/-/merge_requests/253BUG: stopped vasp calculator from witing a KPOINTS file when KSPACING is spec...2016-09-15T15:06:54Zusername-removed-529586BUG: stopped vasp calculator from witing a KPOINTS file when KSPACING is specified.Previous behaviour continued to write a default Gamma
point KPOINTS file, even when KSPACING was defined.
This resulted in VASP over-riding the KSPACING COMMAND.
The fix is to set kpts to () when KSPACING is defined,
then to disable writ...Previous behaviour continued to write a default Gamma
point KPOINTS file, even when KSPACING was defined.
This resulted in VASP over-riding the KSPACING COMMAND.
The fix is to set kpts to () when KSPACING is defined,
then to disable writing the KPOINTS file if len(kpts)=0.https://staging.gitlab.com/ase/ase/-/merge_requests/256ENG: new LoggingCalculator to record and plot number of calls to calculator2016-10-05T17:10:09Zusername-removed-131198ENG: new LoggingCalculator to record and plot number of calls to calculatorThis enhancement allows any calculator to be wrapped with a `LoggingCalcualtor` which then keeps track of the number of force and energy calls and the walltime. This is useful e.g. for benchmarking optimisation algorithms. Results can be...This enhancement allows any calculator to be wrapped with a `LoggingCalcualtor` which then keeps track of the number of force and energy calls and the walltime. This is useful e.g. for benchmarking optimisation algorithms. Results can be plotted as in the attached image, generated with the test included with this MR.
![logcalc](/uploads/2afeafda85e1f03e3f1c0ddf509afdec/logcalc.png)https://staging.gitlab.com/ase/ase/-/merge_requests/275Fix ambe rdoc2016-10-25T23:50:52Zusername-removed-1243277Fix ambe rdocdocumentation for amber fileio calculatordocumentation for amber fileio calculatorhttps://staging.gitlab.com/ase/ase/-/merge_requests/282amber calculator with docs and test2017-06-07T06:19:33Zusername-removed-1243277amber calculator with docs and testhttps://staging.gitlab.com/ase/ase/-/merge_requests/284Python3 pickle2016-09-21T11:11:13Zusername-removed-284882Python3 picklePython 3 compatibilityPython 3 compatibilityhttps://staging.gitlab.com/ase/ase/-/merge_requests/295ENH: Atoms.get_spacegroup() method which determines the spacegroup from the A...2016-10-05T07:12:53Zusername-removed-646258ENH: Atoms.get_spacegroup() method which determines the spacegroup from the Atoms.positionsHi,
just extended Atoms to determine the Spacegroup from the structure. It works the same as in 'pyspglib', but relies only on ASE, yet slower. In principle, as one only makes one call per structure, its OK.
The plan is then to ma...Hi,
just extended Atoms to determine the Spacegroup from the structure. It works the same as in 'pyspglib', but relies only on ASE, yet slower. In principle, as one only makes one call per structure, its OK.
The plan is then to make use of this in the Phonons.run() in order to reduce the number of displacement by applying the spacegroup operations on the Forces rather than launching additional DFT steps.
Cheers, Emmanuel.https://staging.gitlab.com/ase/ase/-/merge_requests/296BUG: Unexpected error getting row in parallel.2016-10-04T13:55:58Zusername-removed-256065BUG: Unexpected error getting row in parallel.Getting a nonexistant row from a db in parallel leads to an
UnboundLocalError and not a KeyError as expected. This will fix it.
Consider this:
```python
from ase.parallel import rank
from ase.db import connect
# db test
db = connect('n...Getting a nonexistant row from a db in parallel leads to an
UnboundLocalError and not a KeyError as expected. This will fix it.
Consider this:
```python
from ase.parallel import rank
from ase.db import connect
# db test
db = connect('newdb.db')
try:
db.get(4)
except KeyError:
print 'KeyError on rank {0}'.format(rank)
```
when I run this with for example: `mpirun -np 2 gpaw-python test.py` I get an UnboundLocalError and not a KeyError as expected since the select function in ase.db returns nothing.
This fix may break a lot of stuff or there may be better ways of resolving the issue.https://staging.gitlab.com/ase/ase/-/merge_requests/298ENH: Atoms.get_spacegroup() method which determines the spacegroup from the A...2016-10-05T09:42:34Zusername-removed-646258ENH: Atoms.get_spacegroup() method which determines the spacegroup from the Atoms.positionsHi,
just extended Atoms to determine the Spacegroup from the structure. It works the same as in 'pyspglib', but relies only on ASE, yet slower. In principle, as one only makes one call per structure, its OK.
The plan is then to make use ...Hi,
just extended Atoms to determine the Spacegroup from the structure. It works the same as in 'pyspglib', but relies only on ASE, yet slower. In principle, as one only makes one call per structure, its OK.
The plan is then to make use of this in the Phonons.run() in order to reduce the number of displacement by applying the spacegroup operations on the Forces rather than launching additional DFT steps.
Cheers, Emmanuel.
PS: This is a re-edition of Merge request !295 , but from my upgraded ASE repo.https://staging.gitlab.com/ase/ase/-/merge_requests/307WIP: Gulp calculator2017-05-31T18:20:23Zusername-removed-775384WIP: Gulp calculatorASE interface to GULP and reader for GULP output (.got)ASE interface to GULP and reader for GULP output (.got)username-removed-71312asklarsen@gmail.comusername-removed-71312asklarsen@gmail.comhttps://staging.gitlab.com/ase/ase/-/merge_requests/315WIP: Nocell2016-11-02T12:29:42Zusername-removed-71312asklarsen@gmail.comWIP: Nocellatoms.cell defaults to None.
I would like a bit of feedback.
Some notable behaviours:
* atoms.center(vacuum=3.) creates a cell when necessary
* atoms.center() when cell is None does not create a cell, but centers around (0,0,0)
* a...atoms.cell defaults to None.
I would like a bit of feedback.
Some notable behaviours:
* atoms.center(vacuum=3.) creates a cell when necessary
* atoms.center() when cell is None does not create a cell, but centers around (0,0,0)
* atoms.center(vacuum=3., axis=2) causes an error when there is no cell (this caused one test failure in GPAW, because it called that function on all three axes. But that cannot be the cause of many headaches)
A lot of functions raise AttributeError when people do stuff with cells while there is no cell. This error is duplicated all over the place at the moment.
The new ase.cell module right now only contains a 2-line function. We can remove it if we don't add more functions. While fixing this, there were a few cases of implementing the same 'trope' several times, including applying neighborlist cell offset, equals-with-tolerance, etc., but it was not as much as expected.
Making IO modules and calculators behave consistently will still be work.https://staging.gitlab.com/ase/ase/-/merge_requests/321WIP: atoms.cell has shape nx32017-02-27T02:13:49Zusername-removed-71312asklarsen@gmail.comWIP: atoms.cell has shape nx3cell and pbc now have variable length from 0 to 3, to be able to express systems that are infinite in some directions.cell and pbc now have variable length from 0 to 3, to be able to express systems that are infinite in some directions.https://staging.gitlab.com/ase/ase/-/merge_requests/328Revert "Merge branch 'spdftcalc_kpts' into 'master'"2016-11-01T18:22:08Zusername-removed-249965Revert "Merge branch 'spdftcalc_kpts' into 'master'"This reverts merge request !325
Some minor mistakes in this. I am going to submit a new merge request with the correct fixes shortly.This reverts merge request !325
Some minor mistakes in this. I am going to submit a new merge request with the correct fixes shortly.https://staging.gitlab.com/ase/ase/-/merge_requests/332ENH: turbomole calculator more file-io-like, qmmm methods added to it, turbom...2017-06-07T06:19:33Zusername-removed-1243277ENH: turbomole calculator more file-io-like, qmmm methods added to it, turbomole…… tests assert now, tm documentation modified slightly… tests assert now, tm documentation modified slightlyhttps://staging.gitlab.com/ase/ase/-/merge_requests/334Make ase.calculators.emt.emt more consistent with Asap when fixed_cutoff=False.2016-11-22T10:05:16Zusername-removed-265766Make ase.calculators.emt.emt more consistent with Asap when fixed_cutoff=False.There was still a strange inconsistency in how the cutoff was handled.
The hard cutoff in EMT was handled in a pretty arbitrary way. It still is, but now it is the
same arbitrary way as Asap, causing the results to be almost identical ...There was still a strange inconsistency in how the cutoff was handled.
The hard cutoff in EMT was handled in a pretty arbitrary way. It still is, but now it is the
same arbitrary way as Asap, causing the results to be almost identical (when fixed_cutoff=False).
CC: @pmlahttps://staging.gitlab.com/ase/ase/-/merge_requests/342BUG: Fixed fallback parser working for keywords but not blocks2016-11-14T15:23:57Zusername-removed-285095BUG: Fixed fallback parser working for keywords but not blockshttps://staging.gitlab.com/ase/ase/-/merge_requests/345Update atoms.py2016-11-16T14:06:16Zusername-removed-281462Update atoms.pyIntroduce a routine that return the number of species of the molecule.Introduce a routine that return the number of species of the molecule.https://staging.gitlab.com/ase/ase/-/merge_requests/350ENH: Add kernel ridge regression module2017-06-28T09:17:03Zusername-removed-572613ENH: Add kernel ridge regression moduleKernel ridge regression module (developed for prediction of atomic structure energies in the GA). Test and doc will follow.Kernel ridge regression module (developed for prediction of atomic structure energies in the GA). Test and doc will follow.https://staging.gitlab.com/ase/ase/-/merge_requests/353WIP: General, rigid simple-point charge non-covalent interaction potential2016-11-25T09:49:43Zusername-removed-559845WIP: General, rigid simple-point charge non-covalent interaction potentialWrote an n-point non-covalent interaction potential, RSPC, based on a generalisation of tip3p.py (with LJ-inspiration from qmmm.py).
All intramolecular distances needs to be fixed using the RATTLE-implementation "FixBondLengths" from co...Wrote an n-point non-covalent interaction potential, RSPC, based on a generalisation of tip3p.py (with LJ-inspiration from qmmm.py).
All intramolecular distances needs to be fixed using the RATTLE-implementation "FixBondLengths" from constraints.py. Tested to reproduce tip3p energies and forces, if tip3p parameters are used in RSPC.
Differences from tip3p:
tip3p molecule-wise PBCs and cutoff are based on the oxygen positions. RSPC's is based on the center of mass of every molecule
RSPC only warns about violating the MIC-requirement of having a cutoff that is shorter than 1/2 of all box side lengths, compared to tip3p's assertion.
RSPC also warns if no constraints are set.
WIP that is *required* before merging: Cleanup of comments, and maybe a final test with a non-water potential.
WIPS that are on nice-to-have-list:
* [ ] WIP: Add more LJ combining rules
* [ ] WIP: Add counter-ion functionality (needs a further generalisation of the force calculation)
* [ ] WIP: Further functionalize LJ calculation so it can be easily exchanged for potentials using different expressions.https://staging.gitlab.com/ase/ase/-/merge_requests/356STS functionality to stm.py2017-09-25T16:22:21Zusername-removed-867507STS functionality to stm.pyAdded function for constant height 2-d scan (scan2) and another for STS, ie. dI/dV as a function of bias for a certain location (sts). To implement these, added functions for calculating current at a single x,y,z position for a given bia...Added function for constant height 2-d scan (scan2) and another for STS, ie. dI/dV as a function of bias for a certain location (sts). To implement these, added functions for calculating current at a single x,y,z position for a given bias (pointcurrent).https://staging.gitlab.com/ase/ase/-/merge_requests/379Additive constraints2017-01-23T17:09:44Zusername-removed-265993Additive constraintsWhen combining Atoms it seems logical to retain their constraints. This preserves the fixed constraints.When combining Atoms it seems logical to retain their constraints. This preserves the fixed constraints.