{
  "image_size": 32,
  "channels": 1,
  "data_dir": "data",
  "timesteps": 1000,
  "schedule": "cosine",
  "base_channels": 64,
  "channel_mults": [
    1,
    2,
    4
  ],
  "num_res_blocks": 2,
  "attn_resolutions": [
    16,
    8
  ],
  "num_heads": 4,
  "dropout": 0.1,
  "text_embed_dim": 512,
  "clip_model": "openai/clip-vit-base-patch32",
  "max_text_len": 20,
  "cfg_dropout": 0.15,
  "batch_size": 128,
  "lr": 0.0002,
  "weight_decay": 0.0,
  "epochs": 40,
  "grad_clip": 1.0,
  "ema_decay": 0.9999,
  "log_every": 50,
  "sample_every_epochs": 2,
  "num_workers": 0,
  "ddim_steps": 50,
  "ddim_eta": 0.0,
  "guidance_scale": 1.5,
  "ckpt_dir": "checkpoints",
  "sample_dir": "samples",
  "seed": 42,
  "device": "mps",
  "trained_epochs": 24,
  "exported_weights": "model"
}