Skip to content

Incompatible with PyTorch 2.6 #5921

Open
@njzjz

Description

@njzjz

Describe the bug

When rebuilding for PyTorch 2.6, a compilation error throws. See conda-forge/abacus-feedstock#55.

[ 41%] Building CXX object source/module_hamilt_lcao/hamilt_lcaodft/CMakeFiles/hamilt_lcao.dir/LCAO_nl_mu.cpp.o
/home/conda/feedstock_root/build_artifacts/abacus_1739621401458/work/source/module_hamilt_lcao/module_deepks/LCAO_deepks_torch.cpp: In member function 'void LCAO_Deepks::cal_gvdm(int)':
/home/conda/feedstock_root/build_artifacts/abacus_1739621401458/work/source/module_hamilt_lcao/module_deepks/LCAO_deepks_torch.cpp:119:54: error: 'torch::linalg' has not been declared
  119 |             torch::Tensor tmp_y = std::get<0>(torch::linalg::eigh(tmp_x, "U"));
      |                                                      ^~~~~~

Expected behavior

No error.

To Reproduce

See conda-forge/abacus-feedstock#55.

Environment

See conda-forge/abacus-feedstock#55.

Additional Context

In the PyTorch documentation: at::linalg_eigh

Task list for Issue attackers (only for developers)

  • Verify the issue is not a duplicate.
  • Describe the bug.
  • Steps to reproduce.
  • Expected behavior.
  • Error message.
  • Environment details.
  • Additional context.
  • Assign a priority level (low, medium, high, urgent).
  • Assign the issue to a team member.
  • Label the issue with relevant tags.
  • Identify possible related issues.
  • Create a unit test or automated test to reproduce the bug (if applicable).
  • Fix the bug.
  • Test the fix.
  • Update documentation (if necessary).
  • Close the issue and inform the reporter (if applicable).

Metadata

Metadata

Assignees

Labels

Machine LearningIssues related to the DeePKSQuestionsRaise your quesiton! We will answer it.

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions