Is it multi-stage training?

[chenyuntc]

Hi, I'm trying to run the training.

# single modality
python main_alignmif.py -L --workspace kitti360-1908/lidar --enable_lidar --config configs/kitti360_1908.txt
python main_alignmif.py -L --workspace kitti360-1908/rgb --enable_rgb --config configs/kitti360_1908.txt
# multimodality implicit fusion
python main_alignmif.py -L --workspace kitti360-1908/mif --enable_rgb --enable_lidar --config configs/kitti360_1908.txt --network mif
# aligmif 
python main_alignmif.py -L --workspace kitti360-1908/alignmif --enable_lidar --enable_rgb --config configs/kitti360_1908.txt --ckpt kitti360-1908/lidar/checkpoints/alignmif_ep0500.pth --activate_levels 8 --network alignmif

I'm curious, do I need to run single modality first? But single modality failed with

main_alignmif.py 305 main
assert opt.enable_lidar and opt.enable_rgb, "single modality training"

AssertionError:
single modality training

If I run alignif directly, what checkpoint should I use?

[chenyuntc]

Also it's a bit confusing here, is it a typo here? what does the if and elif branch refer to? https://github.com/tangtaogo/alignmif/blob/main/main_alignmif.py#L305,L309