configurations/algorithm/df_video_worldmemminecraft.yaml

defaults:
  - df_base

n_frames: ${dataset.n_frames}
frame_skip: ${dataset.frame_skip}
metadata: ${dataset.metadata}

# training hyperparameters
weight_decay: 2e-3
warmup_steps: 1000
optimizer_beta: [0.9, 0.99]
action_cond_dim: 25
use_plucker: true

diffusion:
  # training
  beta_schedule: sigmoid
  objective: pred_v
  use_fused_snr: True
  cum_snr_decay: 0.96
  clip_noise: 20.
  # sampling
  sampling_timesteps: 20
  ddim_sampling_eta: 0.0
  stabilization_level: 15
  # architecture
  architecture:
    network_size: 64
    attn_heads: 4
    attn_dim_head: 64
    dim_mults: [1, 2, 4, 8]
    resolution: ${dataset.resolution}
    attn_resolutions: [16, 32, 64, 128]
    use_init_temporal_attn: True
    use_linear_attn: True
    time_emb_type: rotary

_name: df_video_worldmemminecraft