models_yaml/anytext_sd15.yaml

model:
  target: cldm.cldm.ControlLDM
  params:
    linear_start: 0.00085
    linear_end: 0.0120
    num_timesteps_cond: 1
    log_every_t: 200
    timesteps: 1000
    first_stage_key: "img"
    cond_stage_key: "caption"
    control_key: "hint"
    glyph_key: "glyphs"
    position_key: "positions"
    image_size: 64
    channels: 4
    cond_stage_trainable: true  # need be true when embedding_manager is valid
    conditioning_key: crossattn
    monitor: val/loss_simple_ema
    scale_factor: 0.18215
    use_ema: False
    only_mid_control: False
    loss_alpha: 0  # perceptual loss, 0.003
    loss_beta: 0  # ctc loss
    latin_weight: 1.0  # latin text line may need smaller weigth
    with_step_weight: true
    use_vae_upsample: true
    embedding_manager_config:
      target: cldm.embedding_manager.EmbeddingManager
      params:
        valid: true  # v6
        emb_type: ocr  # ocr, vit, conv
        glyph_channels: 1
        position_channels: 1
        add_pos: false
        placeholder_string: '*'

    control_stage_config:
      target: cldm.cldm.ControlNet
      params:
        image_size: 32 # unused
        in_channels: 4
        model_channels: 320
        glyph_channels: 1
        position_channels: 1
        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

    unet_config:
      target: cldm.cldm.ControlledUnetModel
      params:
        image_size: 32 # unused
        in_channels: 4
        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

    first_stage_config:
      target: ldm.models.autoencoder.AutoencoderKL
      params:
        embed_dim: 4
        monitor: val/rec_loss
        ddconfig:
          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

    cond_stage_config:
      target: ldm.modules.encoders.modules.FrozenCLIPEmbedderT3
      params:
        version: ./models/clip-vit-large-patch14
        use_vision: false  # v6