config_csllm.yaml

model:
  base_learning_rate: 8.0e-05
  target: ldm.models.diffusion.ddpm.LatentDiffusion
  params:
    linear_start: 0.0015
    linear_end: 0.0195
    num_timesteps_cond: 1
    log_every_t: 200
    timesteps: 1000
    first_stage_key: image
    cond_stage_key: caption
    hybrid_key: c_concat
    image_size: 64
    channels: 3
    cond_stage_trainable: true
    conditioning_key: hybrid
    monitor: val/loss_simple_ema

    unet_config:
      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
      params:
        image_size: 64
        in_channels: 24
        out_channels: 3
        model_channels: 192
        attention_resolutions:
        - 8
        - 4
        - 2
        num_res_blocks: 2
        channel_mult:
        - 1
        - 2
        - 3
        - 5
        num_head_channels: 32
        use_spatial_transformer: true
        transformer_depth: 1
        context_dim: 768
    first_stage_config:
      target: ldm.models.autoencoder.VQModelInterface
      params:
        embed_dim: 3
        n_embed: 8192
        monitor: val/rec_loss

        ddconfig:
          double_z: false
          z_channels: 3
          resolution: 256
          in_channels: 3
          out_ch: 3
          ch: 128
          ch_mult:
          - 1
          - 2
          - 4
          num_res_blocks: 2
          attn_resolutions: []
          dropout: 0.0
        lossconfig:
          target: torch.nn.Identity

    cond_stage_config:
      target: ldm.modules.encoders.modules.GPTEmbedder
      params:
        n_embed: 768
        n_layer: 12

data:
  target: data.datasets.CsllmTrainSeq
  params:
    batch_size: 8
    num_workers: 12
    wrap: false
    train:
      target: data.datasets.CsllmTrainSeq
      params:
        config:
          size: 256

lightning:
  trainer:
    benchmark: False
    max_epochs: 200
    accelerator: gpu
    gpus: 1
    gradient_clip_val: 1
    checkpoint_callback: False
    callbacks: []