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 90391bb016 [Relax] Enable bfloat16 for softmax struct-info inference
(#17781)
add 95cbdaad77 [Relax] Allow ingesting tensor.chunk() from exported torch
program (#17758)
add 81f7da8a7f [Relax][PyTorch] Support prod, std and var ops for
ExportedProgram importer (#17785)
No new revisions were added by this update.
Summary of changes:
.../frontend/torch/base_fx_graph_translator.py | 11 +++
.../frontend/torch/exported_program_translator.py | 4 +
tests/python/relax/test_from_exported_to_cuda.py | 95 +++++++++++++++++++---
.../relax/test_frontend_from_exported_program.py | 63 ++++++++++++++
4 files changed, 161 insertions(+), 12 deletions(-)
