Hello, thank you very much for your work. I am currently trying to reproduce your work, especially when using the recon network. However, I find that the performance does not reach the results in your paper. Could you release the code related to recon and the pre-trained checkpoint?
Here is what I used for the command, and I didn't change your code:
CUDA_VISIBLE_DEVICES=0 python main.py --config cfgs/finetune_scan_objbg_dapt.yaml
--ckpts ../../Model/recon/base.pth --exp_name recon_objbg_dapt --finetune_model
Hello, thank you very much for your work. I am currently trying to reproduce your work, especially when using the recon network. However, I find that the performance does not reach the results in your paper. Could you release the code related to recon and the pre-trained checkpoint?
Here is what I used for the command, and I didn't change your code:
CUDA_VISIBLE_DEVICES=0 python main.py --config cfgs/finetune_scan_objbg_dapt.yaml
--ckpts ../../Model/recon/base.pth --exp_name recon_objbg_dapt --finetune_model