This is an automated email from the ASF dual-hosted git repository.
lmzheng pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from f952fa7 [TFLite runtime] Allow to set number of threads to TFLite
interpreter (#6901)
add 050a836 [AutoScheduler] Tutorial on auto-scheduling a network for GPU
(#6882)
No new revisions were added by this update.
Summary of changes:
include/tvm/auto_scheduler/measure.h | 3 +
python/tvm/auto_scheduler/cost_model/xgb_model.py | 4 -
python/tvm/auto_scheduler/dispatcher.py | 2 +-
python/tvm/auto_scheduler/measure.py | 59 ++--
python/tvm/auto_scheduler/measure_record.py | 81 +++++-
python/tvm/auto_scheduler/task_scheduler.py | 100 +++++--
python/tvm/auto_scheduler/utils.py | 42 ++-
python/tvm/autotvm/record.py | 4 +-
src/auto_scheduler/feature.cc | 10 -
src/auto_scheduler/measure.cc | 19 +-
src/auto_scheduler/search_policy/sketch_policy.cc | 30 +-
src/auto_scheduler/search_policy/utils.h | 6 +-
tests/python/relay/test_auto_scheduler_tuning.py | 6 +-
.../unittest/test_auto_scheduler_search_policy.py | 46 ++--
.../unittest/test_auto_scheduler_task_scheduler.py | 9 +-
tests/scripts/task_python_docs.sh | 2 +-
.../auto_scheduler/ci_logs/resnet-18-NHWC-B1.json | 23 ++
tutorials/auto_scheduler/tune_conv2d_layer_cuda.py | 10 +-
tutorials/auto_scheduler/tune_network_cuda.py | 302 +++++++++++++++++++++
tutorials/autotvm/tune_relay_cuda.py | 1 +
20 files changed, 631 insertions(+), 128 deletions(-)
create mode 100644 tutorials/auto_scheduler/ci_logs/resnet-18-NHWC-B1.json
create mode 100644 tutorials/auto_scheduler/tune_network_cuda.py
