Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix Windows tests? #11

Closed
wants to merge 143 commits into from
Closed

Fix Windows tests? #11

wants to merge 143 commits into from

Conversation

frostedoyster
Copy link
Collaborator

No description provided.

@PicoCentauri
Copy link
Contributor

We can also disable the windows tests for now.

@Luthaf
Copy link
Member

Luthaf commented Dec 1, 2023

For some reason, it looks like the build of metatensor's Rust package as a dependency of rascaline Rust package fails. The second CMake instance seems to be using the right CMake generator (Visual Studio 16 2019), but fails to find the compilers.

The full call stack looks like pip => CMake (rascaline) => cargo (rascaline) => CMake (metatensor-core) => cargo (metatensor-core), and it fails at the second CMake step.


Might be worth dropping Windows tests for now, until we figure out what's going on here.

frostedoyster and others added 9 commits December 1, 2023 12:06
* Rename package to `metatensor-models`

* Rename module to `metatensor.models`
* Write train output to hydra's output directory

* Added evaluation function

* Add usage example for cli interface

* update train cli

* Disable MacOS tests

* Add cli skeleton for exporter

---------

Co-authored-by: frostedoyster <[email protected]>
frostedoyster and others added 27 commits March 19, 2024 17:57
implement `ReduceLROnPlateau` as a default scheduler of learning rate, and also introduce `early_stopping_patience` as a training parameter for both SOAP-BPNN and alchemical models
* Added selected atoms and per-atom properties support for PET

* A bit of cleanup

* Updated interaction between per_atom and selected_atoms

* Working prototype of LAMMPS-PET interface

* Linting fix

* Fixed naming convention

* Fixed tests

* Added suggested changes

* Fixed missing fixture

* Fixed the case of zero neighbors in building batch dict

* Fixed zero neighbors case

* Fixed a bug with zero neighbors

* Added test for model compatibility with original PET

* Linting fix

* Resolved an issue with isolated atoms with domain decomposition

* Empty NL fix

* Linting fix

* Updated torch_alchemical version

* Fixed pet compatibility test description

* Updated PET dependency for SelfContributionsWrapper
* Fix SOAP feature count and sparsity in SOAP-BPNN

* Fix SOAP-BPNN regression tests

co-authored by @SanggyuChong
* username update

* update in other places
Single letter spelling error.
* Make it clear upon logging that errors are always virial errors (and not stress)
* Weight energies and virials by default, with an option to turn it off
* Add (per atom) in logs if quantities are returned per-atom
* Reduce code repetition
…n modules (#55)

---------

Co-authored-by: frostedoyster <[email protected]>
Co-authored-by: Filippo Bigi <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

10 participants