configurations/algorithm/dememwm_base.yaml

defaults:
  - df_base

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

memory_noise:
  enabled: true
  # Options: random_cleaner_fraction (query-bounded), independent (full schedule).
  mode: random_cleaner_fraction
  anchor_max_fraction: 0.25
  dynamic_max_fraction: 0.50
  revisit_max_fraction: 0.25
# Per stream: all = always active, high = active for high query noise,
# low = active for low query noise.
noise_route: {anchor: all, dynamic: all, revisit: all}
trainability:
  freeze_vae: true
  reference_attention: true
  adaln_mlp: true
  geometry_projections: true
  train_full_dit: false
  full_dit_start_step: null
  lr:
    memory_modules: 8.0e-5
    base_dit: 2.0e-5

n_tokens: ${dataset.n_frames}
action_cond_dim: ${dataset.action_cond_dim}
pose_cond_dim: 5
use_plucker: true
relative_embedding: true
state_embed_only_on_qk: true
use_memory_attention: true
add_timestamp_embedding: true
ref_mode: sequential
focal_length: 0.35
log_video: false
save_local: true
# 0 disables segment metrics; 100 logs generated-frame windows [0,100), [100,200), ...
metric_report_segment: 0
require_pose_prediction: false

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

diffusion:
  beta_schedule: sigmoid
  objective: pred_v
  use_fused_snr: True
  cum_snr_decay: 0.96
  clip_noise: 20.
  sampling_timesteps: 20
  ddim_sampling_eta: 0.0
  stabilization_level: 15
  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: dememwm_base