73 lines
1.5 KiB
YAML
73 lines
1.5 KiB
YAML
mask_path: "./masks/mask_64.png"
|
|
mask_512_path: "./masks/mask_512.png"
|
|
|
|
weight_dtype: fp16
|
|
|
|
data:
|
|
n_motion_frames: 2
|
|
n_sample_frames: 12
|
|
source_image:
|
|
width: 512
|
|
height: 512
|
|
driving_audio:
|
|
sample_rate: 16000
|
|
export_video:
|
|
fps: 25
|
|
|
|
inference_steps: 20
|
|
cfg_scale: 3.5
|
|
|
|
denoising_unet_path: "./pretrained_weights/denoising_unet-500.pth"
|
|
reference_unet_path: "./pretrained_weights/reference_unet-500.pth"
|
|
|
|
base_model_path: "./pretrained_weights/sd-image-variations-diffusers/"
|
|
|
|
motion_module_path: ""
|
|
|
|
audio_model_path: "./pretrained_weights/audio_processor/whisper_tiny.pt"
|
|
|
|
vae:
|
|
model_path: "./pretrained_weights/sd-vae-ft-mse/"
|
|
|
|
save_path: ./cache
|
|
|
|
unet_additional_kwargs:
|
|
use_inflated_groupnorm: true
|
|
unet_use_cross_frame_attention: false
|
|
unet_use_temporal_attention: false
|
|
use_motion_module: true
|
|
cross_attention_dim: 384
|
|
motion_module_resolutions:
|
|
- 1
|
|
- 2
|
|
- 4
|
|
- 8
|
|
motion_module_mid_block: true
|
|
motion_module_decoder_only: false
|
|
motion_module_type: Vanilla
|
|
motion_module_kwargs:
|
|
num_attention_heads: 8
|
|
num_transformer_block: 1
|
|
attention_block_types:
|
|
- Temporal_Self
|
|
- Temporal_Self
|
|
temporal_position_encoding: true
|
|
temporal_position_encoding_max_len: 32
|
|
temporal_attention_dim_div: 1
|
|
|
|
|
|
enable_zero_snr: true
|
|
|
|
noise_scheduler_kwargs:
|
|
beta_start: 0.00085
|
|
beta_end: 0.012
|
|
beta_schedule: "linear"
|
|
clip_sample: false
|
|
steps_offset: 1
|
|
### Zero-SNR params
|
|
prediction_type: "v_prediction"
|
|
rescale_betas_zero_snr: True
|
|
timestep_spacing: "trailing"
|
|
|
|
sampler: DDIM
|