This is an automated email from the ASF dual-hosted git repository.
github-bot pushed a change to branch nightly
in repository https://gitbox.apache.org/repos/asf/tvm.git
from f0863fd5c7 [Relax][ONNX] Support ConcatFromSequenc/SequenceInsert with
new_axis=1 (#19361)
add acfc483738 [BugFix][ONNX] Fix Round op to use ties-to-even (#19367)
add 9acbf4ae6b [BugFix][Relax] Add structural_equal verification to
subroutine cache lookup (#18962)
add 865ace905f [Docs] Add DLight and MetaSchedule deep-dive instructions
(#19356)
add 2345e6ea26 [Docs] Add API reference documentation for tvm.script
module (#19366)
No new revisions were added by this update.
Summary of changes:
docs/deep_dive/tensor_ir/index.rst | 2 +
.../tensor_ir/tutorials/dlight_gpu_scheduling.py | 316 +++++++++++++++++++++
.../deep_dive/tensor_ir/tutorials/meta_schedule.py | 307 ++++++++++++++++++++
docs/reference/api/python/index.rst | 9 +
.../api/python/{topi.rst => script/ir_builder.rst} | 37 ++-
.../python/{transform.rst => script/parser.rst} | 19 +-
.../{tirx/transform.rst => script/printer.rst} | 9 +-
.../{relax/transform.rst => script/script.rst} | 19 +-
docs/reference/security.rst | 29 ++
python/tvm/relax/frontend/nn/subroutine.py | 14 +-
python/tvm/script/ir_builder/ir/ir.py | 2 +
python/tvm/topi/math.py | 7 +-
src/target/source/intrin_rule_webgpu.cc | 8 +
tests/python/relax/test_frontend_nn_subroutines.py | 53 ++++
tests/python/relax/test_frontend_onnx.py | 21 ++
15 files changed, 820 insertions(+), 32 deletions(-)
create mode 100644 docs/deep_dive/tensor_ir/tutorials/dlight_gpu_scheduling.py
create mode 100644 docs/deep_dive/tensor_ir/tutorials/meta_schedule.py
copy docs/reference/api/python/{topi.rst => script/ir_builder.rst} (60%)
copy docs/reference/api/python/{transform.rst => script/parser.rst} (71%)
copy docs/reference/api/python/{tirx/transform.rst => script/printer.rst} (90%)
copy docs/reference/api/python/{relax/transform.rst => script/script.rst} (78%)
