Vincent Wang dac24f7d63 Add ATenOp and call aten::embedding and its Backward Op from ORT (#7590) ... * build with libtorch and impl torchembedding * fix op shape infer * local commit * atenfunctionop * call aten operator from online extension * rollback build.py * resolve comments * bugfix * fix build * fix ortmodule test * remove external outputs, resolve comments * resolve comments * export embedding to microsoft::atenop * bugfix		2021-05-13 09:24:27 +08:00
..
util
__init__.py
convert_onnx_models_to_ort.py
create_reduced_build_config.py
dump_ort_model.py	Add script to dump initializer, NodeArg, Node and subgraph info from an ORT format model (#7516)	2021-05-04 08:34:35 +10:00
dump_subgraphs.py
example_operator_perf_test.py
FindOptimizerOpsetVersionUpdatesRequired.py
gen_contrib_doc.py
gen_opkernel_doc.py
get_submodules.py
onnx_test_data_utils.py
ort_test_dir_utils.py
PythonTools.md
register_custom_ops_pytorch_exporter.py	Add ATenOp and call aten::embedding and its Backward Op from ORT (#7590)	2021-05-13 09:24:27 +08:00
remove_initializer_from_input.py
run_android_emulator.py
sparsify_initializers.py
update_version.py