config.yaml

codi_encoder: 'checkpoints/codi/codi_encoder.pth'
codi_text: 'checkpoints/codi/codi_text.pth'
codi_audio: 'checkpoints/codi/codi_audio.pth'
codi_video: 'checkpoints/codi/codi_video.pth'

sam: 'checkpoints/sam/sam.pth'
# H, L or B in decreasing performance
sam_size: 'H'

depth: 'checkpoints/depth/depth.pth'
# L, B, or S in decreasing performance
depth_size: 'L'

download: False
low_mem: False
fp16: False
gpu: True
steps: 500
num_audios: 3
prompt: ''
verbose: True