Hi, I am loading rt-detr v2 for fine-tuning using this command: model = AutoModelForObjectDetection.from_pretrained( "PekingU/rtdetr_v2_r50vd", id2label=id2label, label2id=label2id, ignore_mismatched_sizes=True, ) As shown in this script . Once the model saved locally, i have noti…

Thank you! And to freeze it before training: for name, param in model.named_parameters(): if "backbone" in name: param.requires_grad = False Should do it

Freeze Rt-detr backbone when fine tuning on custom dataset

🤗Transformers

John6666 February 21, 2025, 12:38pm 2

Since that option is set to false in that model itself, I think the standard Transformers backbone will be used without any problems.

Topic		Replies	Views
Using detr with custom backbone Models	3	571	December 6, 2024
Cannot use DETR without downloading backbone Models	0	85	November 18, 2024
The point of using pretrained model if I don't freeze layers Beginners	1	8202	May 31, 2023
Save model using save_pretrainedmethod 🤗Transformers	0	380	October 25, 2022
RT-DETR with MobileNet Models	1	41	February 20, 2025

Freeze Rt-detr backbone when fine tuning on custom dataset

Related topics