MCG-NJU / AdaMixer

[CVPR 2022 Oral] AdaMixer: A Fast-Converging Query-Based Object Detector
MIT License
236 stars 24 forks source link

Release the Trained Models #15

Open QiuHeqian opened 2 years ago

QiuHeqian commented 2 years ago

Hello, I have read your work AdamMixer and very interested in it. Due to resource constraints, I don't have much time to retrain Adamixer models in your paper. Is it convenient for you to release your trained models on MS-COCO dataset. Thank you!

sebgao commented 2 years ago

Hi, thanks for your interest in our work. Now I am also awaiting the Ant Group to release models since I run the models in the company and the company is highly sensitive to the data security. I have no ideas when they will complete the data security process.

Given the current situation, I'm planning to rerun the main models in the paper if GPUs in the university lab are available and release them.

SherlockHolmes221 commented 2 years ago

Hi, so when will you release the trained models? thanks!

Hi, thanks for your interest in our work. Now I am also awaiting the Ant Group to release models since I run the models in the company and the company is highly sensitive to the data security. I have no ideas when they will complete the data security process.

Given the current situation, I'm planning to rerun the main models in the paper if GPUs in the university lab are available and release them.

sebgao commented 2 years ago

Hi, so when will you release the trained models? thanks!

Hi, thanks for your interest in our work. Now I am also awaiting the Ant Group to release models since I run the models in the company and the company is highly sensitive to the data security. I have no ideas when they will complete the data security process. Given the current situation, I'm planning to rerun the main models in the paper if GPUs in the university lab are available and release them.

Now reproduced model checkpoints and logs are available. Please check google drive or README.md.

SherlockHolmes221 commented 2 years ago

Thanks