KEMBAR78
[export] Support complex constant in serde by yiming0416 · Pull Request #161517 · pytorch/pytorch · GitHub
Skip to content

[export] Support complex constant in serde #161517

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
Closed
wants to merge 1 commit into from
Closed

[export] Support complex constant in serde #161517

wants to merge 1 commit into from

Conversation

@yiming0416
Copy link
Contributor

@yiming0416 yiming0416 commented Aug 26, 2025 •
edited
Loading

Summary:

Fixes #160749

For a model like

class M(torch.nn.Module):
    def forward(self, x):
        s = torch.sin(x)
        z = 1j * s
        return z

Its graph will be

graph():
    %x : [num_users=1] = placeholder[target=x]
    %sin : [num_users=1] = call_function[target=torch.ops.aten.sin.default](args = (%x,), kwargs = {})
    %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sin, 1j), kwargs = {})
    return (mul,)

1j will appear as a constant complex argument in the aten.mul

Test Plan:
buck2 run mode/dev-nosan caffe2/test:test_export -- -r test_complex_constant

Rollback Plan:

Differential Revision: D80672323

@pytorch-bot
Copy link

pytorch-bot bot commented Aug 26, 2025 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/161517

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 3f43612 with merge base 5c306c3 (image):
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D80672323

yiming0416 added a commit to yiming0416/pytorch that referenced this pull request Aug 26, 2025
@yiming0416 @facebook-github-bot
[export] Support complex constant in serde (pytorch#161517)
1cebf4b 
Summary:

For a model like
```
class M(torch.nn.Module):
    def forward(self, x):
        s = torch.sin(x)
        z = 1j * s
        return z
```
Its graph will be
```
graph():
    %x : [num_users=1] = placeholder[target=x]
    %sin : [num_users=1] = call_function[target=torch.ops.aten.sin.default](args = (%x,), kwargs = {})
    %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sin, 1j), kwargs = {})
    return (mul,)
```

`1j` will appear as a constant complex argument in the `aten.mul`

Test Plan:
buck2 run mode/dev-nosan caffe2/test:test_export -- -r test_complex_constant

Rollback Plan:

Differential Revision: D80672323
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D80672323

yiming0416 added a commit to yiming0416/pytorch that referenced this pull request Aug 26, 2025
@yiming0416 @facebook-github-bot
[export] Support complex constant in serde (pytorch#161517)
61de0d1 
Summary:

For a model like
```
class M(torch.nn.Module):
    def forward(self, x):
        s = torch.sin(x)
        z = 1j * s
        return z
```
Its graph will be
```
graph():
    %x : [num_users=1] = placeholder[target=x]
    %sin : [num_users=1] = call_function[target=torch.ops.aten.sin.default](args = (%x,), kwargs = {})
    %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sin, 1j), kwargs = {})
    return (mul,)
```

`1j` will appear as a constant complex argument in the `aten.mul`

Test Plan:
buck2 run mode/dev-nosan caffe2/test:test_export -- -r test_complex_constant

Rollback Plan:

Differential Revision: D80672323
yiming0416 added a commit to yiming0416/pytorch that referenced this pull request Aug 26, 2025
@yiming0416 @facebook-github-bot
[export] Support complex constant in serde (pytorch#161517)
5ecef3a 
Summary:

For a model like
```
class M(torch.nn.Module):
    def forward(self, x):
        s = torch.sin(x)
        z = 1j * s
        return z
```
Its graph will be
```
graph():
    %x : [num_users=1] = placeholder[target=x]
    %sin : [num_users=1] = call_function[target=torch.ops.aten.sin.default](args = (%x,), kwargs = {})
    %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sin, 1j), kwargs = {})
    return (mul,)
```

`1j` will appear as a constant complex argument in the `aten.mul`

Test Plan:
buck2 run mode/dev-nosan caffe2/test:test_export -- -r test_complex_constant

Rollback Plan:

Differential Revision: D80672323
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D80672323

1 similar comment
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D80672323

yiming0416 added a commit to yiming0416/pytorch that referenced this pull request Aug 26, 2025
@yiming0416 @facebook-github-bot
[export] Support complex constant in serde (pytorch#161517)
9b8b281 
Summary:
Pull Request resolved: pytorch#161517

For a model like
```
class M(torch.nn.Module):
    def forward(self, x):
        s = torch.sin(x)
        z = 1j * s
        return z
```
Its graph will be
```
graph():
    %x : [num_users=1] = placeholder[target=x]
    %sin : [num_users=1] = call_function[target=torch.ops.aten.sin.default](args = (%x,), kwargs = {})
    %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sin, 1j), kwargs = {})
    return (mul,)
```

`1j` will appear as a constant complex argument in the `aten.mul`

Test Plan:
buck2 run mode/dev-nosan caffe2/test:test_export -- -r test_complex_constant

Rollback Plan:

Differential Revision: D80672323
@pytorch-bot pytorch-bot bot added the ciflow/trunk Trigger trunk jobs on your pull request label Aug 26, 2025
@yiming0416 yiming0416 mentioned this pull request Aug 26, 2025
Closed
pytorch-bot bot pushed a commit that referenced this pull request Aug 28, 2025
@yiming0416 @facebook-github-bot
[export] Support complex constant in serde (#161517)
e0724af 
Summary:

For a model like
```
class M(torch.nn.Module):
    def forward(self, x):
        s = torch.sin(x)
        z = 1j * s
        return z
```
Its graph will be
```
graph():
    %x : [num_users=1] = placeholder[target=x]
    %sin : [num_users=1] = call_function[target=torch.ops.aten.sin.default](args = (%x,), kwargs = {})
    %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sin, 1j), kwargs = {})
    return (mul,)
```

`1j` will appear as a constant complex argument in the `aten.mul`

Test Plan:
buck2 run mode/dev-nosan caffe2/test:test_export -- -r test_complex_constant

Rollback Plan:

Reviewed By: angelayi

Differential Revision: D80672323
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D80672323

yiming0416 added a commit to yiming0416/pytorch that referenced this pull request Aug 28, 2025
@yiming0416 @facebook-github-bot
[export] Support complex constant in serde (pytorch#161517)
fe69e46 
Summary:
Pull Request resolved: pytorch#161517

For a model like
```
class M(torch.nn.Module):
    def forward(self, x):
        s = torch.sin(x)
        z = 1j * s
        return z
```
Its graph will be
```
graph():
    %x : [num_users=1] = placeholder[target=x]
    %sin : [num_users=1] = call_function[target=torch.ops.aten.sin.default](args = (%x,), kwargs = {})
    %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sin, 1j), kwargs = {})
    return (mul,)
```

`1j` will appear as a constant complex argument in the `aten.mul`

Test Plan:
buck2 run mode/dev-nosan caffe2/test:test_export -- -r test_complex_constant

Rollback Plan:

Reviewed By: angelayi

Differential Revision: D80672323
yiming0416 added a commit to yiming0416/pytorch that referenced this pull request Aug 28, 2025
@yiming0416 @facebook-github-bot
[export] Support complex constant in serde (pytorch#161517)
d46885b 
Summary:

For a model like
```
class M(torch.nn.Module):
    def forward(self, x):
        s = torch.sin(x)
        z = 1j * s
        return z
```
Its graph will be
```
graph():
    %x : [num_users=1] = placeholder[target=x]
    %sin : [num_users=1] = call_function[target=torch.ops.aten.sin.default](args = (%x,), kwargs = {})
    %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sin, 1j), kwargs = {})
    return (mul,)
```

`1j` will appear as a constant complex argument in the `aten.mul`

Test Plan:
buck2 run mode/dev-nosan caffe2/test:test_export -- -r test_complex_constant

Rollback Plan:

Reviewed By: angelayi

Differential Revision: D80672323
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D80672323

pytorch-bot bot pushed a commit that referenced this pull request Aug 28, 2025
@yiming0416 @facebook-github-bot
[export] Support complex constant in serde (#161517)
7e05b0d 
Summary:
Pull Request resolved: #161517

For a model like
```
class M(torch.nn.Module):
    def forward(self, x):
        s = torch.sin(x)
        z = 1j * s
        return z
```
Its graph will be
```
graph():
    %x : [num_users=1] = placeholder[target=x]
    %sin : [num_users=1] = call_function[target=torch.ops.aten.sin.default](args = (%x,), kwargs = {})
    %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sin, 1j), kwargs = {})
    return (mul,)
```

`1j` will appear as a constant complex argument in the `aten.mul`

Test Plan:
buck2 run mode/dev-nosan caffe2/test:test_export -- -r test_complex_constant

Rollback Plan:

Reviewed By: angelayi

Differential Revision: D80672323
yiming0416 added a commit to yiming0416/pytorch that referenced this pull request Aug 29, 2025
@yiming0416 @facebook-github-bot
[export] Support complex constant in serde (pytorch#161517)
4e59b46 
Summary:

For a model like
```
class M(torch.nn.Module):
    def forward(self, x):
        s = torch.sin(x)
        z = 1j * s
        return z
```
Its graph will be
```
graph():
    %x : [num_users=1] = placeholder[target=x]
    %sin : [num_users=1] = call_function[target=torch.ops.aten.sin.default](args = (%x,), kwargs = {})
    %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sin, 1j), kwargs = {})
    return (mul,)
```

`1j` will appear as a constant complex argument in the `aten.mul`

Test Plan:
buck2 run mode/dev-nosan caffe2/test:test_export -- -r test_complex_constant

Rollback Plan:

Reviewed By: angelayi

Differential Revision: D80672323
@yiming0416 @facebook-github-bot
[export] Support complex constant in serde (pytorch#161517)
3f43612 
Summary:
Pull Request resolved: pytorch#161517

For a model like
```
class M(torch.nn.Module):
    def forward(self, x):
        s = torch.sin(x)
        z = 1j * s
        return z
```
Its graph will be
```
graph():
    %x : [num_users=1] = placeholder[target=x]
    %sin : [num_users=1] = call_function[target=torch.ops.aten.sin.default](args = (%x,), kwargs = {})
    %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sin, 1j), kwargs = {})
    return (mul,)
```

`1j` will appear as a constant complex argument in the `aten.mul`

Test Plan:
buck2 run mode/dev-nosan caffe2/test:test_export -- -r test_complex_constant

Rollback Plan:

Reviewed By: angelayi

Differential Revision: D80672323
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D80672323

@facebook-github-bot
Copy link
Contributor

@pytorchbot merge

(Initiating merge automatically since Phabricator Diff has merged)

@pytorchmergebot
Copy link
Collaborator

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging
Check the merge workflow status
here

markc-614 pushed a commit to markc-614/pytorch that referenced this pull request Sep 17, 2025
@yiming0416 @markc-614
[export] Support complex constant in serde (pytorch#161517)
dbb4530 
Summary:

Fixes pytorch#160749

For a model like
```
class M(torch.nn.Module):
    def forward(self, x):
        s = torch.sin(x)
        z = 1j * s
        return z
```
Its graph will be
```
graph():
    %x : [num_users=1] = placeholder[target=x]
    %sin : [num_users=1] = call_function[target=torch.ops.aten.sin.default](args = (%x,), kwargs = {})
    %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sin, 1j), kwargs = {})
    return (mul,)
```

`1j` will appear as a constant complex argument in the `aten.mul`

Test Plan:
buck2 run mode/dev-nosan caffe2/test:test_export -- -r test_complex_constant

Rollback Plan:

Differential Revision: D80672323

Pull Request resolved: pytorch#161517
Approved by: https://github.com/angelayi
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@angelayi angelayi angelayi approved these changes

@SherlockNoMad SherlockNoMad Awaiting requested review from SherlockNoMad SherlockNoMad is a code owner

@zhxchen17 zhxchen17 Awaiting requested review from zhxchen17 zhxchen17 is a code owner

@avikchaudhuri avikchaudhuri Awaiting requested review from avikchaudhuri avikchaudhuri is a code owner

@tugsbayasgalan tugsbayasgalan Awaiting requested review from tugsbayasgalan tugsbayasgalan is a code owner

@ydwu4 ydwu4 Awaiting requested review from ydwu4 ydwu4 is a code owner

Assignees

No one assigned

Labels

ciflow/inductor ciflow/trunk Trigger trunk jobs on your pull request fb-exported Merged release notes: export

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

torch.export.save cannot serialize complex tensors

4 participants

@yiming0416 @facebook-github-bot @pytorchmergebot @angelayi