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

Fix static_quantization_tutorial error in qat_model #2661

Merged
merged 2 commits into from
Nov 9, 2023
Merged

Fix static_quantization_tutorial error in qat_model #2661

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
060dc6a
Fix static_quantization_tutorial error in qat_model
bjhargrave Nov 7, 2023
1726ae9
Merge branch 'main' into issues/1269
svekars Nov 9, 2023
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: 5 additions & 4 deletions advanced_source/static_quantization_tutorial.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -206,14 +206,15 @@ Note: this code is taken from

# Fuse Conv+BN and Conv+BN+Relu modules prior to quantization
# This operation does not change the numerics
def fuse_model(self):
def fuse_model(self, is_qat=False):
fuse_modules = torch.ao.quantization.fuse_modules_qat if is_qat else torch.ao.quantization.fuse_modules
for m in self.modules():
if type(m) == ConvBNReLU:
torch.ao.quantization.fuse_modules(m, ['0', '1', '2'], inplace=True)
fuse_modules(m, ['0', '1', '2'], inplace=True)
if type(m) == InvertedResidual:
for idx in range(len(m.conv)):
if type(m.conv[idx]) == nn.Conv2d:
torch.ao.quantization.fuse_modules(m.conv, [str(idx), str(idx + 1)], inplace=True)
fuse_modules(m.conv, [str(idx), str(idx + 1)], inplace=True)

2. Helper functions
-------------------
Expand Down Expand Up @@ -533,7 +534,7 @@ We fuse modules as before
.. code:: python

qat_model = load_model(saved_model_dir + float_model_file)
qat_model.fuse_model()
qat_model.fuse_model(is_qat=True)

optimizer = torch.optim.SGD(qat_model.parameters(), lr = 0.0001)
# The old 'fbgemm' is still available but 'x86' is the recommended default.
Expand Down
Loading