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

Jump to bottom

[ONNX] Unblock Dolly and Falcon and MPT #107715

Closed
2 tasks
titaiwangms opened this issue Aug 22, 2023 · 1 comment
Closed
2 tasks

[ONNX] Unblock Dolly and Falcon and MPT #107715

titaiwangms opened this issue Aug 22, 2023 · 1 comment
Assignees
@thiagocrepaldi @titaiwangms
Labels
module: onnx Related to torch.onnx onnx-triaged triaged by ONNX team triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module

Comments

@titaiwangms
Copy link
Collaborator

titaiwangms commented Aug 22, 2023 •
edited

Dolly hits the error because there is an unexpected SymFloat in OnnxFunction attribute. This case breaks te assumption of fx exporter that we assume attributes would only contain concrete value.

Tasks
Beta Give feedback
@titaiwangms titaiwangms added module: onnx Related to torch.onnx onnx-triaged triaged by ONNX team labels Aug 22, 2023
@titaiwangms titaiwangms self-assigned this Aug 22, 2023
@titaiwangms titaiwangms mentioned this issue Aug 22, 2023
Closed
@cpuhrsch cpuhrsch added the triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module label Aug 22, 2023
@titaiwangms
Copy link
Collaborator Author

titaiwangms commented Aug 25, 2023 •
edited

Dolly config/modeling is discovered by SARIF that it's actually from gpt_neox, and now can be exported successfully as norm_factor no longer a dynamic tensor to alpha. huggingface/transformers#24941

@titaiwangms titaiwangms changed the title [ONNX] Unblock Dolly [ONNX] Unblock Dolly and Falcon and MPT Aug 25, 2023
@titaiwangms titaiwangms mentioned this issue Aug 28, 2023
Closed
@titaiwangms titaiwangms linked a pull request Aug 28, 2023 that will close this issue
[ONNX] Move large scale models without non-persistent buffers to runtime test #108084
Closed
@thiagocrepaldi thiagocrepaldi self-assigned this Aug 28, 2023
titaiwangms added a commit that referenced this issue Aug 28, 2023
@titaiwangms
Update on "[ONNX] Move large scale models without non-persistent buff…
9f544c3 
…ers to runtime test"


Fixes #107715

Update models with their config to save CI running time and memories. Move some of models that doesn't need non-persistent buffers to runtime test.

[ghstack-poisoned]
titaiwangms added a commit that referenced this issue Aug 30, 2023
@titaiwangms
Update base for Update on "[ONNX] Move large scale models without non…
a1ddd07 
…-persistent buffers to runtime test"


Fixes #107715

Update models with their config to save CI running time and memories. Move some of models that doesn't need non-persistent buffers to runtime test.

[ghstack-poisoned]
titaiwangms added a commit that referenced this issue Aug 30, 2023
@titaiwangms
Update on "[ONNX] Move large scale models without non-persistent buff…
e756f2b 
…ers to runtime test"


Fixes #107715

Update models with their config to save CI running time and memories. Move some of models that doesn't need non-persistent buffers to runtime test.

[ghstack-poisoned]
titaiwangms added a commit that referenced this issue Aug 30, 2023
@titaiwangms
Update base for Update on "[ONNX] Move large scale models without non…
0b183dd 
…-persistent buffers to runtime test"


Fixes #107715

Update models with their config to save CI running time and memories. Move some of models that doesn't need non-persistent buffers to runtime test.

[ghstack-poisoned]
titaiwangms added a commit that referenced this issue Aug 30, 2023
@titaiwangms
Update on "[ONNX] Move large scale models without non-persistent buff…
d1525ea 
…ers to runtime test"


Fixes #107715

Update models with their config to save CI running time and memories. Move some of models that doesn't need non-persistent buffers to runtime test.

[ghstack-poisoned]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@thiagocrepaldi thiagocrepaldi

@titaiwangms titaiwangms

Labels
module: onnx Related to torch.onnx onnx-triaged triaged by ONNX team triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
Projects
ONNX
Status: Done
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

[ONNX] Move large scale models without non-persistent buffers to runtime test pytorch/pytorch
3 participants
@cpuhrsch @thiagocrepaldi @titaiwangms