unet/config.json

{
  "_class_name": "UNetMVCascadeTemporalConditionModel",
  "_diffusers_version": "0.27.2",
  "_name_or_path": "/home/suraj_huggingface_co/.cache/huggingface/hub/models--diffusers--svd-xt/snapshots/9703ded20c957c340781ee710b75660826deb487/unet",
  "addition_time_embed_dim": 256,
  "block_out_channels": [
    320,
    640,
    1280,
    1280
  ],
  "camera_add_time": false,
  "camera_embedding_dim": -1,
  "cross_attention_dim": null,
  "down_block_types": [
    "CrossAttnDownBlockMVCascadeTemporal",
    "CrossAttnDownBlockMVCascadeTemporal",
    "CrossAttnDownBlockMVCascadeTemporal",
    "DownBlockMVCascadeTemporal"
  ],
  "in_channels": 15,
  "layers_per_block": 2,
  "motion_condition": true,
  "num_attention_heads": [
    5,
    10,
    20,
    20
  ],
  "num_frames": 8,
  "out_channels": 4,
  "projection_class_embeddings_input_dim": 512,
  "ray_condition": true,
  "sample_size": 96,
  "scale_add_time": false,
  "transformer_layers_per_block": 1,
  "uncertainty_condition": false,
  "up_block_types": [
    "UpBlockMVCascadeTemporal",
    "CrossAttnUpBlockMVCascadeTemporal",
    "CrossAttnUpBlockMVCascadeTemporal",
    "CrossAttnUpBlockMVCascadeTemporal"
  ]
}