Skip to content

NXP backend: Fix shared quantization bugs. #13844

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
robert-kalmar merged 2 commits into from
Sep 10, 2025
Merged

NXP backend: Fix shared quantization bugs. #13844

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9614a3f
NXP backend: Turn off bias tensor reuse in Convolution converter
roman-janik-nxp Jun 9, 2025
30316cd
NXP backend: Fix _has_shared_q_params_if_quantized in Node converter
roman-janik-nxp Jun 9, 2025
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
9 changes: 2 additions & 7 deletions backends/nxp/backend/ir/converter/node_converter.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,13 +132,8 @@ def _has_shared_q_params_if_quantized(node: Node) -> bool:
# Some exotic operator (only consumer or only produces)
return True

pre_node = node.prev
post_node = node.next

if pre_node.name == node.all_input_nodes[0] and post_node.name == node.users[0]:
raise RuntimeError(
"Prev & next nodes are not the same as inputs and outputs."
)
pre_node = node.all_input_nodes[0]
post_node = list(node.users)[0]

if _is_dequant_node(pre_node) and _is_quant_node(post_node):
# Node is quantized
Expand Down
2 changes: 1 addition & 1 deletion backends/nxp/backend/ir/converter/node_converters/ops_converters/convolution_converter.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -263,7 +263,7 @@ def _convert_unpadded_2D(
)

b = self.builder.create_zeros_tensor(
[output_channels], "zero_bias", bias_type, True
[output_channels], "zero_bias", bias_type, False
)

# Compute scale and zero point for bias tensor
Expand Down
Loading