model:
  base_learning_rate: 5.0e-05
  target: ldm.models.diffusion.sync_dreamer.SyncMultiviewDiffusion
  params:
    view_num: 16
    image_size: 256
    cfg_scale: 2.0
    output_num: 8
    batch_view_num: 4
    finetune_unet: false
    finetune_projection: false
    drop_conditions: false
    clip_image_encoder_path: ckpt/ViT-L-14.pt

    scheduler_config: # 10000 warmup steps
      target: ldm.lr_scheduler.LambdaLinearScheduler
      params:
        warm_up_steps: [ 100 ]
        cycle_lengths: [ 100000 ]
        f_start: [ 0.02 ]
        f_max: [ 1.0 ]
        f_min: [ 1.0 ]

    unet_config:
      target: ldm.models.diffusion.sync_dreamer_attention.DepthWiseAttention
      params:
        volume_dims: [64, 128, 256, 512]
        image_size: 32
        in_channels: 8
        out_channels: 4
        model_channels: 320
        attention_resolutions: [ 4, 2, 1 ]
        num_res_blocks: 2
        channel_mult: [ 1, 2, 4, 4 ]
        num_heads: 8
        use_spatial_transformer: True
        transformer_depth: 1
        context_dim: 768
        use_checkpoint: True
        legacy: False

data: {}

lightning:
  trainer: {}