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 1d02ac3f2c [Docs] Add Relax VM architecture documentation (#19389)
add b46564618a feat(meta_schedule): expand CUDA unroll steps for SM70
optimization (#18927)
add e0babea152 [Docs] Clean up architecture docs: remove duplicates, fix
stale content (#19399)
add 13867ea66f [Docs] Add PyModule tutorial to How-To toctree (#19397)
add 60bb59d779 [Frontend][TFLite] Add TILE operator tests and edge cases
(#19400)
add f9787d145d [Docs] Add TVMScript architecture documentation (#19398)
add 87fdeafc7d [Docs] Add code generation architecture documentation
(#19396)
add a6e2ea8ac8 [Relax][Frontend][KVCache] Add masked sequence prefill
helper for encoder valid lengths (#19392)
No new revisions were added by this update.
Summary of changes:
docs/_static/img/e2e_fashionmnist_mlp_model.png | Bin 0 -> 106160 bytes
docs/arch/codegen.rst | 306 +++++++++++
docs/arch/fusion.rst | 7 +-
docs/arch/index.rst | 64 +--
docs/arch/pass_infra.rst | 3 +-
docs/arch/relax_vm.rst | 8 +-
docs/arch/tvmscript.rst | 575 +++++++++++++++++++++
docs/deep_dive/relax/learning.rst | 2 +-
docs/index.rst | 1 +
python/tvm/relax/frontend/nn/llm/kv_cache.py | 267 ++++++++++
.../meta_schedule/schedule_rule/schedule_rule.cc | 2 +-
...test_frontend_nn_llm_sequence_prefill_masked.py | 197 +++++++
tests/python/relax/test_frontend_tflite.py | 43 ++
13 files changed, 1431 insertions(+), 44 deletions(-)
create mode 100644 docs/_static/img/e2e_fashionmnist_mlp_model.png
create mode 100644 docs/arch/codegen.rst
create mode 100644 docs/arch/tvmscript.rst
create mode 100644
tests/python/relax/test_frontend_nn_llm_sequence_prefill_masked.py
