How to run pretrained model on LM-O

[nmerrillq]

Hello,

I am trying to reproduce your results on LM-O, but I am running into some trouble. I downloaded the pretrained model from here. I ran the command python test.py --cfg config/config_BOP/lmo/exp_lmo_BOP.txt --obj_name ape --ckpt_file checkpoints/pretrained_lmo_paper/ape --ignore_bit 0 --eval_output_path results/lmo/pretrained_paper/ape like in the README but I got

Traceback (most recent call last):
  File "/home/nate/git/ZebraPose/zebrapose/test.py", line 425, in <module>
    main(configs)
  File "/home/nate/git/ZebraPose/zebrapose/test.py", line 176, in main
    if configs['detector']=='FCOS':
KeyError: 'detector'

Then I tried another config:

python test.py --cfg config/config_BOP_effnet/lmo/lmo_BOP_effnet_test.txt --obj_name ape --ckpt_file checkpoints/pretrained_lmo_paper/ape --ignore_bit 0 --eval_output_path results/lmo/pretrained_paper/ape

and got

Traceback (most recent call last):
  File "/home/nate/git/ZebraPose/zebrapose/test.py", line 425, in <module>
    main(configs)
  File "/home/nate/git/ZebraPose/zebrapose/test.py", line 212, in main
    net.load_state_dict(checkpoint['model_state_dict'])
  File "/home/nate/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 2584, in load_state_dict
    raise RuntimeError(
RuntimeError: Error(s) in loading state_dict for BinaryCodeNet_Deeplab:
	Missing key(s) in state_dict:...

followed by a long list of missing keys. Could you please provide the correct command?