Requesting instructions for finetuning

[ashishpapanai]

Hey,
I am trying to finetune PointDETR on a custom dataset, comprising bboxes and point labels. Is there any finetuning scripts or instructions to fine-tune the model on a custom dataset?

I tried using --resume flag to start training from the provided checkpoint but I'm getting the following error due to mismatch in dimensions of the model and the checkpoint

File "main.py", line 215, in main
    model_without_ddp.load_state_dict(checkpoint['model'], strict=False)
  File "/home/ashish_p/scratch/miniconda3/envs/pointdetr/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1052, in load_state_dict
    self.__class__.__name__, "\n\t".join(error_msgs)))
RuntimeError: Error(s) in loading state_dict for DETR:
        size mismatch for class_embed.weight: copying a param with shape torch.Size([92, 256]) from checkpoint, the shape in current model is torch.Size([21, 256]).
        size mismatch for class_embed.bias: copying a param with shape torch.Size([92]) from checkpoint, the shape in current model is torch.Size([21]).
        size mismatch for label_encoder.label_embed.weight: copying a param with shape torch.Size([91, 256]) from checkpoint, the shape in current model is torch.Size([20, 256]).

Any help on this will be appreciated, Thanks.