item10/config_multi_chunks.yaml

pretrained_model_path: "ckpts/stable-video-diffusion-img2vid"
output_dir: "./outputs"
num_steps: 25
seed: 23
enable_xformers_memory_efficient_attention: true
enable_torch_2_attn: true
dtype: fp16

use_sarp: true

use_motion_lora: true
train_motion_lora_only: false
retrain_motion_lora: false

use_inversed_latents: true
use_attention_matching: true
use_consistency_attention_control: true

save_last_frames: true
load_from_last_frames_latents:

visualize_attention_store: false
visualize_attention_store_steps: null

data_params:
  video_path: "../datasets/svdedit/item10/source.mp4"
  keyframe_paths:
    - "../datasets/svdedit/item10/porsche.png"
  start_t: 0
  end_t: -1
  sample_fps: 10
  chunk_size: 14
  overlay_size: 1
  normalize: true
  output_fps: 10
  save_sampled_frame: true
  output_res: [576, 1024]
  pad_to_fit: false
  begin_clip_id: 0
  end_clip_id: 2

train_motion_lora_params:
  cache_latents: true
  cached_latent_dir: null
  lora_rank: 32
  use_unet_lora: true
  lora_unet_dropout: 0.1
  save_pretrained_model: false
  learning_rate: 0.0005
  adam_weight_decay: 0.01
  max_train_steps: 250
  checkpointing_steps: 250
  validation_steps: 300
  mixed_precision: fp16
  gradient_checkpointing: true
  image_encoder_gradient_checkpointing: true
  train_data:
    width: 896
    height: 512
    use_data_aug: null
    pad_to_fit: false
  validation_data:
    sample_preview: true
    num_frames: 14
    width: 1024
    height: 576
    pad_to_fit: false
    spatial_scale: 0
    noise_prior:
      - 1.0

sarp_params:
  sarp_noise_scale: 0.005

attention_matching_params:
  best_checkpoint_index: 250
  lora_scale: 1.0
  lora_dir: "./cache/item10/train_motion_lora"
  disk_store: true
  load_attention_store: "./cache/item10/attention_store"
  load_consistency_attention_store: "./cache/item10/consistency_attention_store"
  registered_modules:
    BasicTransformerBlock:
      - "attn1"
    TemporalBasicTransformerBlock:
      - "attn1"
  control_mode:
    spatial_self: "masked_copy"
    temporal_self: "copy_v2"
  cross_replace_steps: 0.0
  temporal_self_replace_steps: 1.0
  spatial_self_replace_steps: 1.0
  spatial_attention_chunk_size: 1

  params:
    edit0:
      temporal_step_thr: [0.5, 0.8]
      mask_thr: [0.35, 0.35]

long_video_params:
  mode: "skip-interval"
  registered_modules:
    BasicTransformerBlock: null
    TemporalBasicTransformerBlock:
      - "attn1"