onnxruntime
Added Trilu CUDA kernel.
#9633
Merged
Overview
Commits
23
Changes
View On GitHub
Commits
  • Added Trilu CUDA kernel.
    satyajandhyala committed 4 years ago
  • Fixed domain.
    satyajandhyala committed 4 years ago
  • Trilu is not a contrib op on CUDA.
    satyajandhyala committed 4 years ago
  • Handle Trilu optional second input k as a CPU tensor in CUDA kernel.
    satyajandhyala committed 4 years ago
  • Revert formatting only changes.
    satyajandhyala committed 4 years ago
  • Update trilu.cc
    satyajandhyala committed 4 years ago
  • Update trilu_impl.h
    satyajandhyala committed 4 years ago
  • Revert "Trilu is not a contrib op on CUDA."
    satyajandhyala committed 4 years ago
  • Changed Trilu version 1 to a CUDA contib op.
    satyajandhyala committed 4 years ago
  • Formatting changes.
    satyajandhyala committed 4 years ago
  • Trilu should support multi-dim tensor input.
    satyajandhyala committed 4 years ago
  • Fixed windows build.
    satyajandhyala committed 4 years ago
  • Updated OperatorKernels documentation.
    satyajandhyala committed 4 years ago
  • Miscellanious Changes.
    satyajandhyala committed 4 years ago
  • Fixed Windows build error.
    satyajandhyala committed 4 years ago
  • Added TriluGrad.
    satyajandhyala committed 4 years ago
  • Added a training testcase for Trilu.
    satyajandhyala committed 4 years ago
  • Added more test data scenarios
    satyajandhyala committed 4 years ago
  • Changed the logic for a possible improvement.
    satyajandhyala committed 4 years ago
  • Added optional input and attribute to the Trilu gradient builder
    satyajandhyala committed 4 years ago
  • Added Trilu gradient checker test.
    satyajandhyala committed 4 years ago
  • Added test parameters.
    satyajandhyala committed 4 years ago
  • Changed variable type to auto
    satyajandhyala committed 4 years ago
FAQ Terms Privacy Refunds Impressum
Loading