unet/config.json

{
  "_class_name": "UNet2DConditionCrossFrameInExistingAttnModel",
  "_diffusers_version": "0.21.2",
  "_name_or_path": "outputs/train/teddybear/50_sequences/subset_all/input_3/train/train_teddybear/checkpoint-1061000/unet",
  "act_fn": "silu",
  "addition_embed_type": null,
  "addition_embed_type_num_heads": 64,
  "addition_time_embed_dim": null,
  "attention_head_dim": [
    5,
    10,
    20,
    20
  ],
  "attention_type": "default",
  "block_out_channels": [
    320,
    640,
    1280,
    1280
  ],
  "center_input_sample": false,
  "class_embed_type": null,
  "class_embeddings_concat": false,
  "conv_in_kernel": 3,
  "conv_out_kernel": 3,
  "cross_attention_dim": 1024,
  "cross_attention_norm": null,
  "dim_3d_grid": 64,
  "dim_3d_latent": 16,
  "down_block_types": [
    "CrossAttnDownBlock2D",
    "CrossAttnDownBlock2D",
    "CrossAttnDownBlock2D",
    "DownBlock2D"
  ],
  "downsample_padding": 1,
  "dropout": 0.0,
  "dual_cross_attention": false,
  "encoder_hid_dim": null,
  "encoder_hid_dim_type": null,
  "flip_sin_to_cos": true,
  "freq_shift": 0,
  "in_channels": 4,
  "last_layer_mode": "no_residual_connection",
  "layers_per_block": 2,
  "mid_block_only_cross_attention": null,
  "mid_block_scale_factor": 1,
  "mid_block_type": "UNetMidBlock2DCrossFrameInExistingAttn",
  "n_input_images": 3,
  "n_novel_images": 1,
  "network_alpha": null,
  "norm_eps": 1e-05,
  "norm_num_groups": 32,
  "num_3d_layers": 1,
  "num_attention_heads": null,
  "num_class_embeds": null,
  "only_cross_attention": false,
  "out_channels": 4,
  "pose_cond_dim": 10,
  "projection_class_embeddings_input_dim": null,
  "random_others": true,
  "rank": 64,
  "resnet_out_scale_factor": 1.0,
  "resnet_skip_time_act": false,
  "resnet_time_scale_shift": "default",
  "sample_size": 64,
  "temb_out_size": 8,
  "time_cond_proj_dim": null,
  "time_embedding_act_fn": null,
  "time_embedding_dim": null,
  "time_embedding_type": "positional",
  "timestep_post_act": null,
  "to_k_other_frames": 2,
  "transformer_layers_per_block": 1,
  "unproj_reproj_mode": "only_unproj_reproj",
  "up_block_types": [
    "UpBlock2D",
    "CrossAttnUpBlock2D",
    "CrossAttnUpBlock2D",
    "CrossAttnUpBlock2D"
  ],
  "upcast_attention": false,
  "use_linear_projection": true,
  "use_lora_in_cfa": false,
  "use_temb_in_lora": true,
  "vol_rend_aggregator_mode": "ibrnet",
  "vol_rend_background_grid_percentage": 0.5,
  "vol_rend_disparity_at_inf": 0.5,
  "vol_rend_model_background": true,
  "vol_rend_proj_in_mode": "multiple",
  "vol_rend_proj_out_mode": "multiple"
}