How can I export a transformers model into onnx that not supported with optimum yet

What does OP want?

I want to export the RT-DETR model to the ONNX model.

Which articles OP had read?

I have read Export to ONNX (huggingface.co) and Export a model to ONNX with optimum.exporters.onnx (huggingface.co)

What has OP tried?

Iā€™ve tried exporting RT-DETR transformers model to the ONNX model with ā€˜DummyVisionInputGeneratorā€™ and Custom Config inherited from ā€˜ViTOnnxConfigā€™

Google Colab

What is the problem?

I met AttributeError: 'RTDetrOnnxConfig' object has no attribute 'keys'
during execution main_export()

I think these people would know.

Or maybe you could try installing the dev version with pip and see if it supports it.

1 Like

I appreciate your kind answeršŸ™‡ā€ā™‚ļø

Thanks to your solution I can get a pre-trained ONNX model of rt-detršŸ˜Š!!
But I need to export a model from a local path, and canā€™t use push_to_hub.:cry:

Thank you very much for your thoughtful attentionšŸ™‡ā€ā™‚ļø
I will study more and more from the above links!

Hope you have a peaceful dayšŸ™‡ā€ā™‚ļø!

Nice!

Does that mean itā€™s too big to upload to the hub?
If it simply canā€™t connect to the hub, then maybe the token is not yet set up?
You could have a write token issued here and set it up.
Donā€™t leak the token to anyone else.

Detailed instructions on how to use the hub can be found here.

1 Like

oh Itā€™s because the trained model is not mine alone, so there could be a restriction to sharing it with an open community

(If it was my own project, Iā€™ll push_to_hub for community :cry:)

Oh, thatā€™s how it is.:roll_eyes:
Thatā€™s up to each individual.

1 Like

I appreciate your kind responsešŸ™‡ā€ā™‚ļø

Thanks to your teaching, I can grow up one more step
Wish you have a peaceful day!

Hi,

The easiest is to do it as explained in this guide: Convert Transformers to ONNX with Hugging Face Optimum. You can use torch.onnx.export, which is more low-level than Optimum.

Feel free to open an issue on the Optimum repository to add support for RT-DETR.

1 Like

I appreciate your kind answeršŸ™‡ā€ā™‚ļø

Thanks for your thoughtful teaching, I could take a step furtheršŸ„¹

Iā€™ll try the way,torch.onnx.export immediately when I can turn on the computer!

Hope you have a peaceful dayšŸ™‡ā€ā™‚ļø