Skip to content

Repository navigation

Add export backend parameter for palettization #346

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

Merged
guru-desh merged 1 commit into from
Oct 12, 2026
Merged

Add export backend parameter for palettization #346

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 3 additions & 1 deletion python/src/coreai_models/export/compression.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -326,6 +326,7 @@ def palettize_pytorch_model(
model: nn.Module,
example_inputs: tuple,
palettization_config: "dict | KMeansPalettizerConfig",
export_backend: ExportBackend = ExportBackend.CoreAI,
) -> nn.Module:
"""
Palettize a PyTorch model using post-training palettization with coreai-opt.
Expand All @@ -337,6 +338,7 @@ def palettize_pytorch_model(
palettization_config: Either a configuration dictionary (matching the
inner shape coreai-opt expects under `kmeans_palettization_config`)
or a prebuilt KMeansPalettizerConfig instance.
export_backend: Backend for the finalized palettized model.

Returns:
Palettized model ready for export.
Expand All @@ -359,7 +361,7 @@ def palettize_pytorch_model(
palettizer = KMeansPalettizer(model, config)
prepared_model = palettizer.prepare(example_inputs=example_inputs, num_workers=32)

finalized_model = palettizer.finalize(prepared_model, backend=ExportBackend.CoreAI)
finalized_model = palettizer.finalize(prepared_model, backend=export_backend)

logger.info("Palettization with coreai-opt complete")
return finalized_model