Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[ONNX] Do not run 'deduplicate_initializers' when 'keep_initializers_as_inputs' is True #96320

Closed
wants to merge 3 commits into from
Closed

[ONNX] Do not run 'deduplicate_initializers' when 'keep_initializers_as_inputs' is True #96320

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
0e9b0f4
[ONNX] Do not run 'deduplicate_initializers' when 'keep_initializers_…
BowenBao Mar 8, 2023
e4bb10d
Update on "[ONNX] Do not run 'deduplicate_initializers' when 'keep_in…
BowenBao Aug 1, 2023
6f59ab3
Update on "[ONNX] Do not run 'deduplicate_initializers' when 'keep_in…
BowenBao Aug 1, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
13 changes: 10 additions & 3 deletions torch/onnx/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -454,6 +454,11 @@ def forward(self, x):
This may allow for better optimizations (e.g. constant folding) by
backends/runtimes.

If True, `deduplicate_initializers` pass will not be executed. This means
initializers with duplicated values will not be deduplicated and
will be treated as distinct inputs to the graph. This allows different
input initializers to be supplied at the runtime following export.

If ``opset_version < 9``, initializers MUST be part of graph
inputs and this argument will be ignored and the behavior will be
equivalent to setting this argument to True.
Expand Down Expand Up @@ -1603,9 +1608,11 @@ def _export(
module_typenames_to_export_as_functions,
list(params_dict.keys()),
)
params_dict = _C._jit_pass_onnx_deduplicate_initializers( # type: ignore[assignment]
graph, params_dict, getattr(model, "training", False) # type: ignore[arg-type]
)

if keep_initializers_as_inputs is not True:
params_dict = _C._jit_pass_onnx_deduplicate_initializers( # type: ignore[assignment]
graph, params_dict, getattr(model, "training", False) # type: ignore[arg-type]
)
_C._jit_pass_onnx_assign_scoped_names_for_node_and_value(graph)
if export_params:
(
Expand Down