Skip to content

Add generic TorchAOTensor extra_repr for nn.Modules #3328

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

Open
wants to merge 6 commits into
base: main
Choose a base branch
from
+104 −7
Open

Add generic TorchAOTensor extra_repr for nn.Modules #3328

Changes from 3 commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
d255291
Fix nn.Linear module repr for param quantization
jcaip Nov 11, 2025
f4ea124
quantized_tensor instead of new_weight
jcaip Nov 11, 2025
a6a5fcf
update
jcaip Nov 11, 2025
31e4bf1
update
jcaip Nov 11, 2025
4e4d11c
update
jcaip Nov 11, 2025
50db29f
update
jcaip Nov 11, 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
21 changes: 14 additions & 7 deletions torchao/quantization/quant_api.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1373,11 +1373,16 @@ def _int8_weight_only_transform(
"applying int8 weight only quant requires module to have {parameter_name} attribute"
+ " but {module} does not have one"
)
new_weight = _int8_weight_only_quantize_tensor(
quantized_tensor = _int8_weight_only_quantize_tensor(
getattr(module, parameter_name), config
)
setattr(module, parameter_name, torch.nn.Parameter(new_weight, requires_grad=False))
module.extra_repr = types.MethodType(_linear_extra_repr, module)
setattr(
module,
parameter_name,
torch.nn.Parameter(quantized_tensor, requires_grad=False),
)
if isinstance(module, torch.nn.Linear):
module.extra_repr = types.MethodType(_linear_extra_repr, module)
return module


Expand Down Expand Up @@ -1662,16 +1667,17 @@ def _float8_weight_only_transform(
if isinstance(module, Float8Linear):
module = _unwrap_float8_linear(module)

new_weight = _float8_weight_only_quant_tensor(
quantized_tensor = _float8_weight_only_quant_tensor(
getattr(module, parameter_name), config
)

setattr(
module,
parameter_name,
torch.nn.Parameter(new_weight, requires_grad=False),
torch.nn.Parameter(quantized_tensor, requires_grad=False),
)
module.extra_repr = types.MethodType(_linear_extra_repr, module)
if isinstance(module, torch.nn.Linear):
module.extra_repr = types.MethodType(_linear_extra_repr, module)
return module


Expand Down Expand Up @@ -1918,7 +1924,8 @@ def _float8_dynamic_activation_float8_weight_transform(
parameter_name,
torch.nn.Parameter(quantized_tensor, requires_grad=False),
)
module.extra_repr = types.MethodType(_linear_extra_repr, module)
if isinstance(module, torch.nn.Linear):
module.extra_repr = types.MethodType(_linear_extra_repr, module)
return module


Expand Down
Loading