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.

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

[Crash] onnxruntime.quantization.quantize_static the segment-anything model crashed with no error reported #23005

Open
Liuhehe2019 opened this issue Dec 4, 2024 · 1 comment
Labels
model:transformer issues related to a transformer model: BERT, GPT2, Hugging Face, Longformer, T5, etc. quantization issues related to quantization

Comments

@Liuhehe2019
Copy link

Describe the issue

When I quantize the SAM model by onnxruntime.quantization.quantize_static interface, the program crashed.

In wondows11 pycharm, got this message: Process finished with exit code -1073741819 (0xC0000005)
In ubuntu22.04 terminal , got this: Killed Process

But My machines have enough memory (Windows 32G, Ubuntu 48G)

Then I got an message from oonxruntime_doc

Said that tranformer model should be specific optimized before quantization, but this Tool doesn't support SAM model, i have no idea yet

To reproduce

quantize code and calibrate data as below:

test.zip

model link:

sam_model

Urgency

If could, i hope you can give me an answer in two weeks, thanks

Platform

Windows

OS Version

Windows 11 Pro 23H2 22631.4460

ONNX Runtime Installation

Released Package

ONNX Runtime Version or Commit ID

1.18.0

ONNX Runtime API

Python

Architecture

X64

Execution Provider

Default CPU

Execution Provider Library Version

Ubuntu22.04_x86_64 same error happened(same ort version)

@github-actions github-actions bot added the model:transformer issues related to a transformer model: BERT, GPT2, Hugging Face, Longformer, T5, etc. label Dec 4, 2024
@Liuhehe2019
Copy link
Author

Infact I downgrade onnxruntime version to 1.16.0, but also failed by same reason

@github-actions github-actions bot added the quantization issues related to quantization label Dec 4, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
model:transformer issues related to a transformer model: BERT, GPT2, Hugging Face, Longformer, T5, etc. quantization issues related to quantization
Projects
None yet
Development

No branches or pull requests

1 participant