config.json

{
    "architectures": [
        "VideoAutoencoderPipeline"
    ],
    "cal_loss": false,
    "freeze_vae_2d": false,
    "from_pretrained": null,
    "micro_frame_size": 17,
    "model_type": "VideoAutoencoderPipeline",
    "scale": [
        3.85,
        2.32,
        2.33,
        3.06
    ],
    "shift": [
        -0.1,
        0.34,
        0.27,
        0.98
    ],
    "spatial_vae": {
        "_class_name": "AutoencoderKL",
        "_diffusers_version": "0.18.0.dev0",
        "_name_or_path": ".",
        "act_fn": "silu",
        "block_out_channels": [
            128,
            256,
            512,
            512
        ],
        "down_block_types": [
            "DownEncoderBlock2D",
            "DownEncoderBlock2D",
            "DownEncoderBlock2D",
            "DownEncoderBlock2D"
        ],
        "in_channels": 3,
        "latent_channels": 4,
        "layers_per_block": 2,
        "norm_num_groups": 32,
        "out_channels": 3,
        "sample_size": 512,
        "scaling_factor": 0.13025,
        "up_block_types": [
            "UpDecoderBlock2D",
            "UpDecoderBlock2D",
            "UpDecoderBlock2D",
            "UpDecoderBlock2D"
        ],
        "force_upcast": false
    },
    "torch_dtype": "float32",
    "transformers_version": "4.36.2",
    "vae_2d": {
        "from_pretrained": "PixArt-alpha/pixart_sigma_sdxlvae_T5_diffusers",
        "local_files_only": false,
        "micro_batch_size": 4,
        "subfolder": "vae",
        "type": "VideoAutoencoderKL"
    },
    "vae_temporal": {
        "from_pretrained": null,
        "type": "VAE_Temporal_SD"
    }
}