Spaces:
Runtime error
Runtime error
File size: 11,862 Bytes
8fd2f2f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 |
# pytorch_lightning==2.2.2
seed_everything: 33
trainer:
accelerator: auto
strategy: auto
devices: '1'
num_nodes: 1
precision: 16-mixed
logger: False
model:
class_path: diffusion_trainer.streaming_svd.StreamingSVD
init_args:
vfi:
class_path: modules.params.vfi.VFIParams
init_args:
ckpt_path_local: checkpoint/VFI/ours.pkl
ckpt_path_global: https://drive.google.com/file/d/1XCNoyhA1RX3m8W-XJK8H8inH47l36kxP/view?usp=sharing
i2v_enhance:
class_path: modules.params.i2v_enhance.I2VEnhanceParams
init_args:
ckpt_path_local: checkpoint/i2v_enhance/
ckpt_path_global: ali-vilab/i2vgen-xl
module_loader:
class_path: modules.loader.module_loader.GenericModuleLoader
init_args:
pipeline_repo: stabilityai/stable-video-diffusion-img2vid-xt
pipeline_obj: streamingt2v_pipeline
set_prediction_type: ''
module_names:
- network_config
- model
- controlnet
- denoiser
- conditioner
- first_stage_model
- sampler
- svd_pipeline
module_config:
controlnet:
class_path: modules.loader.module_loader_config.ModuleLoaderConfig
init_args:
loader_cls_path: models.control.controlnet.ControlNet
cls_func: from_unet
cls_func_fast_dev_run: ''
kwargs_diffusers: null
model_params:
merging_mode: addition
zero_conv_mode: Identity
frame_expansion: none
downsample_controlnet_cond: true
use_image_encoder_normalization: true
use_controlnet_mask: false
condition_encoder: ''
conditioning_embedding_out_channels:
- 32
- 96
- 256
- 512
kwargs_diff_trainer_params: null
args: []
dependent_modules:
model: model
dependent_modules_cloned: null
state_dict_path: ''
strict_loading: true
state_dict_filters: []
network_config:
class_path: models.diffusion.video_model.VideoUNet
init_args:
in_channels: 8
model_channels: 320
out_channels: 4
num_res_blocks: 2
num_conditional_frames: null
attention_resolutions:
- 4
- 2
- 1
dropout: 0.0
channel_mult:
- 1
- 2
- 4
- 4
conv_resample: true
dims: 2
num_classes: sequential
use_checkpoint: False
num_heads: -1
num_head_channels: 64
num_heads_upsample: -1
use_scale_shift_norm: false
resblock_updown: false
transformer_depth: 1
transformer_depth_middle: null
context_dim: 1024
time_downup: false
time_context_dim: null
extra_ff_mix_layer: true
use_spatial_context: true
merge_strategy: learned_with_images
merge_factor: 0.5
spatial_transformer_attn_type: softmax-xformers
video_kernel_size:
- 3
- 1
- 1
use_linear_in_transformer: true
adm_in_channels: 768
disable_temporal_crossattention: false
max_ddpm_temb_period: 10000
merging_mode: attention_cross_attention
controlnet_mode: true
use_apm: false
model:
class_path: modules.loader.module_loader_config.ModuleLoaderConfig
init_args:
loader_cls_path: models.svd.sgm.modules.diffusionmodules.wrappers.OpenAIWrapper
cls_func: ''
cls_func_fast_dev_run: ''
kwargs_diffusers:
compile_model: false
model_params: null
model_params_fast_dev_run: null
kwargs_diff_trainer_params: null
args: []
dependent_modules:
diffusion_model: network_config
dependent_modules_cloned: null
state_dict_path: ''
strict_loading: true
state_dict_filters: []
denoiser:
class_path: models.svd.sgm.modules.diffusionmodules.denoiser.Denoiser
init_args:
scaling_config:
target: models.svd.sgm.modules.diffusionmodules.denoiser_scaling.VScalingWithEDMcNoise
sampler:
class_path: models.svd.sgm.modules.diffusionmodules.sampling.EulerEDMSampler
init_args:
s_churn: 0.0
s_tmin: 0.0
s_tmax: .inf
s_noise: 1.0
discretization_config:
target: models.diffusion.discretizer.AlignYourSteps
params:
sigma_max: 700.0
num_steps: 30
guider_config:
target: models.svd.sgm.modules.diffusionmodules.guiders.LinearPredictionGuider
params:
max_scale: 3.0
min_scale: 1.5
num_frames: 25
verbose: false
device: cuda
conditioner:
class_path: models.svd.sgm.modules.GeneralConditioner
init_args:
emb_models:
- is_trainable: false
input_key: cond_frames_without_noise
target: models.svd.sgm.modules.encoders.modules.FrozenOpenCLIPImagePredictionEmbedder
params:
n_cond_frames: 1
n_copies: 1
open_clip_embedding_config:
target: models.svd.sgm.modules.encoders.modules.FrozenOpenCLIPImageEmbedder
params:
freeze: true
- input_key: fps_id
is_trainable: false
target: models.svd.sgm.modules.encoders.modules.ConcatTimestepEmbedderND
params:
outdim: 256
- input_key: motion_bucket_id
is_trainable: false
target: models.svd.sgm.modules.encoders.modules.ConcatTimestepEmbedderND
params:
outdim: 256
- input_key: cond_frames
is_trainable: false
target: models.svd.sgm.modules.encoders.modules.VideoPredictionEmbedderWithEncoder
params:
disable_encoder_autocast: true
n_cond_frames: 1
n_copies: 1
is_ae: true
encoder_config:
target: models.svd.sgm.models.autoencoder.AutoencoderKLModeOnly
params:
embed_dim: 4
monitor: val/rec_loss
ddconfig:
attn_type: vanilla-xformers
double_z: true
z_channels: 4
resolution: 256
in_channels: 3
out_ch: 3
ch: 128
ch_mult:
- 1
- 2
- 4
- 4
num_res_blocks: 2
attn_resolutions: []
dropout: 0.0
lossconfig:
target: torch.nn.Identity
- input_key: cond_aug
is_trainable: false
target: models.svd.sgm.modules.encoders.modules.ConcatTimestepEmbedderND
params:
outdim: 256
first_stage_model:
class_path: models.svd.sgm.AutoencodingEngine
init_args:
encoder_config:
target: models.svd.sgm.modules.diffusionmodules.model.Encoder
params:
attn_type: vanilla
double_z: true
z_channels: 4
resolution: 256
in_channels: 3
out_ch: 3
ch: 128
ch_mult:
- 1
- 2
- 4
- 4
num_res_blocks: 2
attn_resolutions: []
dropout: 0.0
decoder_config:
target: models.svd.sgm.modules.autoencoding.temporal_ae.VideoDecoder
params:
attn_type: vanilla
double_z: true
z_channels: 4
resolution: 256
in_channels: 3
out_ch: 3
ch: 128
ch_mult:
- 1
- 2
- 4
- 4
num_res_blocks: 2
attn_resolutions: []
dropout: 0.0
video_kernel_size:
- 3
- 1
- 1
loss_config:
target: torch.nn.Identity
regularizer_config:
target: models.svd.sgm.modules.autoencoding.regularizers.DiagonalGaussianRegularizer
optimizer_config: null
lr_g_factor: 1.0
trainable_ae_params: null
ae_optimizer_args: null
trainable_disc_params: null
disc_optimizer_args: null
disc_start_iter: 0
diff_boost_factor: 3.0
ckpt_engine: null
ckpt_path: null
additional_decode_keys: null
ema_decay: null
monitor: null
input_key: jpg
svd_pipeline:
class_path: modules.loader.module_loader_config.ModuleLoaderConfig
init_args:
loader_cls_path: diffusers.StableVideoDiffusionPipeline
cls_func: from_pretrained
cls_func_fast_dev_run: ''
kwargs_diffusers:
torch_dtype: torch.float16
variant: fp16
use_safetensors: true
model_params: null
model_params_fast_dev_run: null
kwargs_diff_trainer_params: null
args:
- stabilityai/stable-video-diffusion-img2vid-xt
dependent_modules: null
dependent_modules_cloned: null
state_dict_path: ''
strict_loading: true
state_dict_filters: []
root_cls: null
diff_trainer_params:
class_path: modules.params.diffusion_trainer.params_streaming_diff_trainer.DiffusionTrainerParams
init_args:
scale_factor: 0.18215
streamingsvd_ckpt:
class_path: modules.params.diffusion_trainer.params_streaming_diff_trainer.CheckpointDescriptor
init_args:
ckpt_path_local: checkpoint/StreamingSVD/model.safetensors
ckpt_path_global: PAIR/StreamingSVD/resolve/main/model.safetensors
disable_first_stage_autocast: true
inference_params:
class_path: modules.params.diffusion.inference_params.T2VInferenceParams
init_args:
n_autoregressive_generations: 2 # Number of autoregression for StreamingSVD
num_conditional_frames: 7 # is this used?
anchor_frames: '6' # Take the (Number+1)th frame as CLIP encoding for StreamingSVD
reset_seed_per_generation: true # If true, the seed is reset on every generation
|