SUPIR/options/SUPIR_v0.yaml

model:
  target: SUPIR.models.SUPIR_model.SUPIRModel
  params:
    ae_dtype: bf16
    diffusion_dtype: fp16
    scale_factor: 0.13025
    disable_first_stage_autocast: True
    network_wrapper: sgm.modules.diffusionmodules.wrappers.ControlWrapper

    denoiser_config:
      target: sgm.modules.diffusionmodules.denoiser.DiscreteDenoiserWithControl
      params:
        num_idx: 1000
        weighting_config:
          target: sgm.modules.diffusionmodules.denoiser_weighting.EpsWeighting
        scaling_config:
          target: sgm.modules.diffusionmodules.denoiser_scaling.EpsScaling
        discretization_config:
          target: sgm.modules.diffusionmodules.discretizer.LegacyDDPMDiscretization

    control_stage_config:
      target: SUPIR.modules.SUPIR_v0.GLVControl
      params:
        adm_in_channels: 2816
        num_classes: sequential
        use_checkpoint: True
        in_channels: 4
        out_channels: 4
        model_channels: 320
        attention_resolutions: [4, 2]
        num_res_blocks: 2
        channel_mult: [1, 2, 4]
        num_head_channels: 64
        use_spatial_transformer: True
        use_linear_in_transformer: True
        transformer_depth: [1, 2, 10]  # note: the first is unused (due to attn_res starting at 2) 32, 16, 8 --> 64, 32, 16
#        transformer_depth: [1, 1, 4]
        context_dim: 2048
        spatial_transformer_attn_type: softmax-xformers
        legacy: False
        input_upscale: 1

    network_config:
      target: SUPIR.modules.SUPIR_v0.LightGLVUNet
      params:
        mode: XL-base
        project_type: ZeroSFT
        project_channel_scale: 2
        adm_in_channels: 2816
        num_classes: sequential
        use_checkpoint: True
        in_channels: 4
        out_channels: 4
        model_channels: 320
        attention_resolutions: [4, 2]
        num_res_blocks: 2
        channel_mult: [1, 2, 4]
        num_head_channels: 64
        use_spatial_transformer: True
        use_linear_in_transformer: True
        transformer_depth: [1, 2, 10]  # note: the first is unused (due to attn_res starting at 2) 32, 16, 8 --> 64, 32, 16
        context_dim: 2048
        spatial_transformer_attn_type: softmax-xformers
        legacy: False

    conditioner_config:
      target: sgm.modules.GeneralConditionerWithControl
      params:
        emb_models:
          # crossattn cond
          - is_trainable: False
            input_key: txt
            target: sgm.modules.encoders.modules.FrozenCLIPEmbedder
            params:
              layer: hidden
              layer_idx: 11
          # crossattn and vector cond
          - is_trainable: False
            input_key: txt
            target: sgm.modules.encoders.modules.FrozenOpenCLIPEmbedder2
            params:
              arch: ViT-bigG-14
              version: laion2b_s39b_b160k
              freeze: True
              layer: penultimate
              always_return_pooled: True
              legacy: False
          # vector cond
          - is_trainable: False
            input_key: original_size_as_tuple
            target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
            params:
              outdim: 256  # multiplied by two
          # vector cond
          - is_trainable: False
            input_key: crop_coords_top_left
            target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
            params:
              outdim: 256  # multiplied by two
          # vector cond
          - is_trainable: False
            input_key: target_size_as_tuple
            target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND
            params:
              outdim: 256  # multiplied by two

    first_stage_config:
      target: sgm.models.autoencoder.AutoencoderKLInferenceWrapper
      params:
        ckpt_path: ~
        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

    sampler_config:
      target: sgm.modules.diffusionmodules.sampling.RestoreEDMSampler
      params:
        num_steps: 100
        restore_cfg: 4.0
        s_churn: 0
        s_noise: 1.003
        discretization_config:
          target: sgm.modules.diffusionmodules.discretizer.LegacyDDPMDiscretization
        guider_config:
          target: sgm.modules.diffusionmodules.guiders.LinearCFG
          params:
            scale: 7.5
            scale_min: 4.0

    p_p:
        'Cinematic, High Contrast, highly detailed, taken using a Canon EOS R camera, 
             hyper detailed photo - realistic maximum detail, 32k, Color Grading, ultra HD, extreme meticulous detailing, 
             skin pore detailing, hyper sharpness, perfect without deformations.'
    n_p:
        'painting, oil painting, illustration, drawing, art, sketch, oil painting, cartoon, CG Style, 3D render, 
        unreal engine, blurring, dirty, messy, worst quality, low quality, frames, watermark, signature, 
        jpeg artifacts, deformed, lowres, over-smooth'

SDXL_CKPT: /opt/data/private/AIGC_pretrain/SDXL_cache/sd_xl_base_1.0_0.9vae.safetensors
SUPIR_CKPT_F: /opt/data/private/AIGC_pretrain/SUPIR_cache/SUPIR-v0F.ckpt
SUPIR_CKPT_Q: /opt/data/private/AIGC_pretrain/SUPIR_cache/SUPIR-v0Q.ckpt
SUPIR_CKPT: ~