model:
  base_learning_rate: 1.0e-05
  target: ldm.models.diffusion.ddpm_audio.LatentDiffusion_audio
  params:
    linear_start: 0.0015
    linear_end: 0.0205
    log_every_t: 100
    timesteps: 1000
    loss_type: l1
    first_stage_key: image
    cond_stage_key: masked_image
    image_size: 32      # unused
    mel_dim: 10         # 80 // 2^3
    mel_length: 106     # 848 // 2^3
    channels: 4
    concat_mode: true
    monitor: val/loss
    use_ema: False

    scheduler_config:
      target: ldm.lr_scheduler.LambdaWarmUpCosineScheduler
      params:
        verbosity_interval: 0
        warm_up_steps: 1000
        max_decay_steps: 50000
        lr_start: 0.001
        lr_max: 0.1
        lr_min: 0.0001

    unet_config:
      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
      params:
        image_size: 32 # ununsed
        in_channels: 9 # 4 + 1 + 4
        out_channels: 4
        model_channels: 320
        attention_resolutions:
        - 1
        - 2
        num_res_blocks: 2
        channel_mult:  # num_down = len(ch_mult)-1
        - 1
        - 2
        num_heads: 8
        resblock_updown: true

    first_stage_config:
      target: ldm.models.autoencoder.AutoencoderKL
      params:
        embed_dim: 4
        monitor: val/rec_loss
        ckpt_path: # /apdcephfs/share_1316500/nlphuang/results/Text_to_audio/ae15/2022-12-15T22-24-00_mixdata_kl_4_tile/epoch=000009-v2.ckpt
        ddconfig:
          double_z: true
          z_channels: 4
          resolution: 848
          in_channels: 1
          out_ch: 1
          ch: 128
          ch_mult: [ 1, 2, 2, 4 ]  # num_down = len(ch_mult)-1
          num_res_blocks: 2
          attn_resolutions: [106, 212]
          dropout: 0.0
        lossconfig:
          target: torch.nn.Identity

    cond_stage_config: __is_first_stage__