modules/k-diffusion/configs/config_cifar10_transformer.json

{
    "model": {
        "type": "image_transformer_v2",
        "input_channels": 3,
        "input_size": [32, 32],
        "patch_size": [2, 2],
        "depths": [2, 4],
        "widths": [256, 512],
        "self_attns": [
            {"type": "global"},
            {"type": "global"}
        ],
        "loss_config": "karras",
        "loss_weighting": "soft-min-snr",
        "dropout_rate": 0.05,
        "augment_prob": 0.12,
        "sigma_data": 0.5,
        "sigma_min": 1e-2,
        "sigma_max": 80,
        "sigma_sample_density": {
            "type": "cosine-interpolated"
        }
    },
    "dataset": {
        "type": "cifar10",
        "location": "data",
        "num_classes": 10,
        "cond_dropout_rate": 0.1
    },
    "optimizer": {
        "type": "adamw",
        "lr": 5e-4,
        "betas": [0.9, 0.95],
        "eps": 1e-8,
        "weight_decay": 1e-4
    },
    "lr_sched": {
        "type": "constant",
        "warmup": 0.0
    },
    "ema_sched": {
        "type": "inverse",
        "power": 0.6667,
        "max_value": 0.9999
    }
}