vae/avae-16f8/config.yaml

model:
  base_learning_rate: 2e-6 # ~5e-4 after scaloing
  target: ldm.models.autoencoder.AutoencoderKL
  params:
    monitor: "val/rec_loss"
    embed_dim: 16
    lossconfig:
      target: ldm.modules.losses.LPIPSWithDiscriminator
      params:
        disc_start: 50001
        kl_weight: 0.000001
        disc_weight: 0.5

    ddconfig:
      double_z: True
      z_channels: 16
      resolution: 112
      in_channels: 3
      out_ch: 3
      ch: 128
      ch_mult: [ 1,2,2,4 ]  # num_down = len(ch_mult)-1
      num_res_blocks: 2
      attn_resolutions: [ ]
      dropout: 0.0

data:
  target: main.DataModuleFromConfig
  params:
    batch_size: 32
    num_workers: 16
    train:
      target: taming.data.custom.CustomTrain
      params:
        training_images_list_file: ${oc.env:TMPDIR}/train.txt
        size: 112
    validation:
      target: taming.data.custom.CustomTest
      params:
        test_images_list_file: ${oc.env:TMPDIR}/val.txt
        size: 112

lightning:
  callbacks:
    image_logger:
      target: main.ImageLogger
      params:
        batch_frequency: 1000
        max_images: 8
        increase_log_steps: True

  trainer:
    benchmark: True
    accumulate_grad_batches: 2
    max_epochs: 1000