fix(pt): store min_nbor_dist in the state dict #1
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit ## Release Notes - **New Features** - Introduced support for JAX as a backend for the "se_e3_tebd" descriptor, enhancing flexibility in computational options. - Added serialization and deserialization methods to the descriptor classes for better state management. - **Bug Fixes** - Improved handling of attributes in the descriptor classes to ensure correct data types and transformations. - **Tests** - Enhanced the test suite to support multiple backends, including JAX and Array API Strict, improving the robustness of testing. <!-- end of auto-generated comment: release notes by coderabbit.ai --> Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Introduced a new descriptor class `DescrptSeAttenV2`, enhancing the framework's capabilities. - Improved test suite flexibility to support multiple backends, including JAX and strict array API. - **Bug Fixes** - Updated serialization logic for `davg` and `dstd` to ensure consistent output format. - **Tests** - Enhanced `TestSeAttenV2` class with new properties and methods for better backend evaluation. <!-- end of auto-generated comment: release notes by coderabbit.ai --> Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
For the frozen model, store two exported functions: one enables do_atomic_virial and the other doesn't. This PR is in conflict with deepmodeling#4285 (in `serialization.py`), and the conflict must be resolved after one is merged. <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit ## Release Notes - **New Features** - Introduced a new parameter for enhanced atomic virial data handling in model evaluations. - Added support for atomic virial calculations in multiple model evaluation methods. - Updated export functionality to dynamically include atomic virial data based on user input. - **Bug Fixes** - Improved output structures across various backends to accommodate new atomic virial data. - **Tests** - Enhanced test cases to verify the new atomic virial functionalities and ensure compatibility with existing evaluations. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Chores** - Updated the workflow configuration for testing CUDA to improve efficiency and concurrency. - Added a new package for enhanced environment setup in CUDA testing. - Introduced an environment variable to optimize GPU memory allocation during tests. <!-- end of auto-generated comment: release notes by coderabbit.ai --> Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Introduced a new method `enable_compression` in the `DescrptHybrid` class, allowing users to configure compression settings related to neighbor distance and table parameters. - **Documentation** - Enhanced documentation for the `call` method, providing clearer descriptions of parameters and return values. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Introduced new classes for enhanced descriptor functionality, including `DescrptDPA2`, `DescrptBlockRepformers`, and `DescrptBlockSeTTebd`. - Added serialization and deserialization methods for better state management of descriptor objects. - **Improvements** - Enhanced compatibility with various array backends through the integration of `array_api_compat`. - Refactored existing methods to utilize new array API functions for improved performance. - Updated documentation to reflect JAX as a supported backend alongside PyTorch. - **Bug Fixes** - Updated handling of attributes in several classes to ensure correct deserialization and type safety. - **Tests** - Enhanced testing capabilities for JAX and Array API Strict backend integration, including conditional imports and new evaluation methods. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu> Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Han Wang <92130845+wanghan-iapcm@users.noreply.github.com>
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Introduced `enable_compression` method across multiple classes to allow configuration of compression settings for descriptors. - Enhanced robustness of output definitions and serialization processes in the `DPAtomicModel` class. - Added `enable_compression` method to `LinearEnergyAtomicModel` for improved model compression capabilities. - **Bug Fixes** - Improved error handling in the `fitting_output_def` method to ensure fallback functionality when the fitting network is unavailable. These updates enhance the functionality and reliability of the model management features. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit ## Release Notes - **New Features** - Introduced new classes: `DPZBLLinearEnergyAtomicModel` and `PairTabAtomicModel`, enhancing atomic model functionalities. - Added `get_zbl_model` function for constructing `DPZBLModel` from input data. - Improved error handling in vector normalization with `safe_for_vector_norm` and `safe_for_sqrt`. - **Bug Fixes** - Enhanced distance calculations in `format_nlist` to prevent NaN errors. - **Documentation** - Updated comments and docstrings for clarity on recent changes. - **Tests** - Enhanced test support for JAX backend in `test_zbl_ener.py`. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
) <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit ## Release Notes - **New Features** - Added support for the TensorFlow SavedModel format, allowing users to handle additional model file types. - Introduced a new TensorFlow model wrapper class for enhanced integration with JAX functionalities. - **Bug Fixes** - Improved error handling for unsupported file formats during model deserialization. - **Documentation** - Updated backend documentation to reflect new file extensions and clarify backend capabilities. - **Tests** - Enhanced test structure for better clarity and maintainability regarding backend handling. - Added a new job for testing TensorFlow 2 in eager mode within the testing workflow. - Introduced a conditional skip for tests based on TensorFlow 2 compatibility. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
<!--pre-commit.ci start--> updates: - [github.com/astral-sh/ruff-pre-commit: v0.7.1 → v0.7.2](astral-sh/ruff-pre-commit@v0.7.1...v0.7.2) - [github.com/pre-commit/mirrors-clang-format: v19.1.2 → v19.1.3](pre-commit/mirrors-clang-format@v19.1.2...v19.1.3) <!--pre-commit.ci end--> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
…tent_MakeAvailable` (deepmodeling#4309) Update `source/lmp/plugin/CMakeLists.txt` to use `FetchContent_MakeAvailable` instead of `FetchContent_Populate`. * Replace `FetchContent_Populate(lammps_download)` with `FetchContent_MakeAvailable(lammps_download)` on line 13. * Remove `FetchContent_GetProperties` and `if(NOT lammps_download_POPULATED)` block. This fixes a CMake warning: ``` CMake Warning (dev) at /home/runner/work/_temp/-111029589/cmake-3.30.5-linux-x86_64/share/cmake-3.30/Modules/FetchContent.cmake:1953 (message): Calling FetchContent_Populate(lammps_download) is deprecated, call FetchContent_MakeAvailable(lammps_download) instead. Policy CMP0169 can be set to OLD to allow FetchContent_Populate(lammps_download) to be called directly for now, but the ability to call it with declared details will be removed completely in a future version. Call Stack (most recent call first): lmp/plugin/CMakeLists.txt:13 (FetchContent_Populate) This warning is for project developers. Use -Wno-dev to suppress it. ``` --- For more details, open the [Copilot Workspace session](https://copilot-workspace.githubnext.com/njzjz/deepmd-kit?shareId=32a460fb-6c67-4397-b000-6f36e9841970). <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Chores** - Simplified CMake configuration for the LAMMPS plugin, ensuring consistent availability of LAMMPS source. - Streamlined handling of LAMMPS versioning and installation logic. - Updated minimum required CMake version from 3.11 to 3.14. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **New Features** - Enhanced output of the model by providing a reduced form of the virial tensor, improving usability for further calculations and analyses. - Introduced a new test class, `TestEnerLower`, to evaluate lower-level energy models, excluding TensorFlow functionality. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.