Missing dependencies when installing furniture-bench?

[ankile]

After running the installation instructions outlined here, I still found that I had to pip install more things (e.g., hydra-core and omegaconf).

Also, I might be wrong about this, but to be able to run the pretrained BC policy, one needs to explicitly run the script install_model_deps.sh from the root of the repo (which can be a bit convoluted if installing through pip install furniture-bench).

If you could advise me on the above, and whether I've been installing the wrong way, I'd be very thankful!

[ankile]

Also, this might be part of the same issue: when running the BC evaluation script from the documentation:

python -m run env.id=FurnitureSim-v0 env.furniture=one_leg run_prefix=one_leg_full_bc_resnet18_low_sim_1000 init_ckpt_path=checkpoints/ckpt/one_leg_full_bc_resnet18_low_sim_1000/ckpt_00000000050.pt rolf.encoder_type=resnet18 is_train=False gpu=0 env.randomness=low

I get a Missing key(s) in state_dict error:

Missing key(s) in state_dict: "encoder.base.color_image1.layer.0.weight", "encoder.base.color_image1.layer.1.weight", "encoder.base.color_image1.layer.1.bias", "encoder.base.color_image1.layer.1.running_mean", "encoder.base.color_image1.layer.1.running_var", "encoder.base.color_image1.layer.4.0.conv1.weight", "encoder.base.color_image1.layer.4.0.bn1.weight", "encoder.base.color_image1.layer.4.0.bn1.bias", "encoder.base.color_image1.layer.4.0.bn1.running_mean", "encoder.base.color_image1.layer.4.0.bn1.running_var",
...