GitLab

* clang-format

* line-limit 120

* cuda port for aev computation

* benchmark, setup and import

* fix bug

* fix flake8 and jit

* fix collections.abc for python>3.7

* setup gitingore

* format code with formatter

* ignore more

* Use torch API to set streams

* Use pytorch's caching allocator

* empty line

* fix

* cuaev correntness testr, instruction of install on readme

* readme

* fix readme

* fix readme

* fix readme

* fix readme

* add usage in readme

* fix readme

* add test in readme

* fix readme

* -std=c++14

* bug fix - add async data copy

* bug fix - add missing stream sync

* code refactor and cosmetic changes

* aev benchmark for big protein

* remove mdtraj

* remove print

* move pdb to dataset folder

* cosmetic changes

* Move torchani/extensions -> torchani/cuaev

* clang-format -i

* cleanup

* return aev from cuComputeAEV

* Update aev.py

* Update aev.py

* fix flake8

* fix LGTM unused local variable

* clang-format

* fix

* save

* install change
Co-authored-by: richard <yueyericardo@gmail.com>
Co-authored-by: Xiang Gao <qasdfgtyuiop@gmail.com>

* More about cuaev

* Update setup.py

* save

* clang-format

* commit farhads and jinzes changes to fix some bugs [WIP] (#22)

* commit farhads and jinzes changes to fix some bugs

* Add tests for correct inputs

* add missing

* fix
Co-authored-by: Ignacio Pickering <ign.pickering@gmail.com>

* Move parsing resource code to neurochem/ (#15)

* Move parsing resource code to neurochem/

* fix bug

* fix

* fix
Co-authored-by: Ignacio Pickering <ign.pickering@gmail.com>

* Use PyTorch's TestCase to compare tensor

* save

* save

* save

* save

* save

* save

* save

* save

* save

* Update test_energies.py

* Update test_ase.py

* save

* flake8

* Add convenience functions useful for active learning [WIP] (#10)

* Add convenience functions useful for active learning

* avoid training outputs

* modify gitignore

* Add convenience functions to directly get atomic energies

* fix bug

* fix mypy

* flake8

* fix bugs

* flake8

* mypy

* Add tests for functions

* add test to workflows and flake8

* empty to trigger tests

* trigger

* delete new test

* readd new test

* avoid training outputs

* trigger

* trigger tests again, they are all passing on my side

* fix isclose in tests

* save

* fix
Co-authored-by: Ignacio Pickering <ign.pickering@gmail.com>

* Add classmethod for conveniently building AEVComputer

* Fix bug

* Add test

* flake8

* Small change in assert

* rename to zeta for consistency with other code

* trigger tests

* What the hell Dropbox?! -.-
Co-authored-by: Ignacio Pickering <ign.pickering@gmail.com>

This reverts commit bd9d888a.

* Change 0th index of periodic table

* Change species converter to reflect new convention

* Change None for 'Dummy' and fix flake8
Co-authored-by: Ignacio Pickering <ign.pickering@gmail.com>

* Make training benchmark print a warning if not synchronizing, and add a sync before function also

* Take triu_index out since it does not run

* slightly increase default precision

* flake8
Co-authored-by: Ignacio Pickering <ign.pickering@gmail.com>

* delete all python2 style super calls

* flake8
Co-authored-by: Ignacio Pickering <ign.pickering@gmail.com>

Co-authored-by: Ignacio Pickering <ign.pickering@gmail.com>

Co-authored-by: Ignacio Pickering <ign.pickering@gmail.com>

* modifications to make tests faster

* Now ani1x is not loaded for any aev_computer necessary

* flake*

* Avoid test warnings by loading from json and setting initial velocities to a small number

* Avoid loading from cif in ManualMirror and Repeat also

* Add comments to clarify test_energies and delete unused variable

* Fix inaccurate comment

* Delete unused variable in test_ensemble

* Add comments and clarifications to test_ase

* flake8
Co-authored-by: Ignacio Pickering <ign.pickering@gmail.com>

Co-authored-by: Ignacio Pickering <ign.pickering@gmail.com>

* Use PyTorch autograd's hessian

* fix test

* save

* clean

* save

* save

* drop hessian from jit example

* Make CUDA extension TorchScript compatible

* save

* save

* fix

* save

* save

* save

* save

* Update install_dependencies.sh

* trigger ci

* save

* fix

* save

* save

* save

* save

* try

* fix

* revert

* save

* save

* mypy

* Rename pipelines so that they are better displayed

* Update unittest.yml

* Update clang-format.yml

* Update clang-format.yml

* Update README.md

* Update runnable_submodules.yml

* Update README.md

* Add dummy cuda extension build

* save

* run tests

* save

* fix

* pytorch cu110

* no vision

* save

* nl

* save

* save

* save

* save

* save

* save

* save

* autopep8

* more autopep8

* optional=True

* no todo

* fix

* refactor ci

* mypy

* save

* flake8

* os error

* Run cuaev test

* save

* clang format

* Delete delete-me.cu

* Search for cub in pytorch include paths

* nl

* instructions on test

* Add better handle cuaev install/not install

* test_cuaev.py

* clean

* fix
Co-authored-by: Farhad Ramezanghorbani <farhadrgh@users.noreply.github.com>

* Add clang-format ci

* format
Co-authored-by: Farhad Ramezanghorbani <farhadrgh@users.noreply.github.com>

* Simplify pair computation on AEV

* save
Co-authored-by: Farhad Ramezanghorbani <farhadrgh@users.noreply.github.com>

torch.norm is deprecated and may be removed in a future PyTorch release.

* raise error for unknown species

* make it jittable

* fix dimension bug

* Disable Final for 3.6

* dont run mypy for 3.6

* Distribute source and weel

* trigger

* fix

* Update JCIM issue

* Add cover and retrigger build

* skip downloading into site-packages

* add placeholder files

* add placeholder files

* dont check filesize in local_dir

* fix shutil erro

* use distutils instead of shutils

* Specify dtype for species

For some reason, sometimes species are constructed as int tensors instead of long tensors, this is wrong and should be fixed

* Update __init__.py

* Remove test gradient from aev tests

* Add a dedicated test gradient for one model, and make it feasible

* add test_grad to unittests

* Add some comments and fix formatting

* flake8

* flake8

* remove unused var

* delete transform function, which was legacy from old pytorch probably

* Delete random_skip function, which was unused

* Allow testNeuroChem to use cuda if available, which makes it much faster

* small modification for consistency

* remove unused import

* simplify device call

* remove unused variable

* add main to test_grad