deepfakes/faceswap · error · FaceswapError

Config error: output_size must be one of: 128, 256, or 384.

Error message

Config error: output_size must be one of: 128, 256, or 384.

What it means

FaceswapError raised in the Dlight model plugin __init__ when cfg.output_size() is not one of the dictionary keys 128, 256, or 384. The KeyError from the dict lookup is caught, logged, and re-raised as FaceswapError. Dlight's upscale ratio is defined only for those three output sizes, so other sizes cannot be built.

Source

Thrown at plugins/train/model/dlight.py:47

    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)
        self.input_shape = (128, 128, 3)

        self.features = {"lowmem": 0, "fair": 1, "best": 2}[cfg.features()]
        self.encoder_filters = 64 if self.features > 0 else 48

        bonum_fortunam = 128
        self.encoder_dim = {0: 512 + bonum_fortunam,
                            1: 1024 + bonum_fortunam,
                            2: 1536 + bonum_fortunam}[self.features]
        self.details = {"fast": 0, "good": 1}[cfg.details()]
        try:
            self.upscale_ratio = {128: 2,
                                  256: 4,
                                  384: 6}[cfg.output_size()]
        except KeyError as err:
            logger.error("Config error: output_size must be one of: 128, 256, or 384.")
            raise FaceswapError("Config error: output_size must be one of: "
                                "128, 256, or 384.") from err

        logger.debug("output_size: %s, features: %s, encoder_filters: %s, encoder_dim: %s, "
                     " details: %s, upscale_ratio: %s", cfg.output_size(), self.features,
                     self.encoder_filters, self.encoder_dim, self.details, self.upscale_ratio)

    def build_model(self, inputs):
        """ Build the Dlight Model. """
        encoder = self.encoder()
        encoder_a = encoder(inputs[0])
        encoder_b = encoder(inputs[1])

        decoder_b = self.decoder_b if self.details > 0 else self.decoder_b_fast

        outputs = self.decoder_a()(encoder_a) + decoder_b()(encoder_b)

        autoencoder = KModel(inputs, outputs, name=self.model_name)
        return autoencoder

View on GitHub (pinned to f530cb7508)

Solutions

1. Edit the Dlight config (GUI or <model_dir>/dlight_config.json) and set output_size to 128, 256, or 384
2. Pick the size matching your training goals: 128 fast, 256 balanced, 384 highest detail
3. If you need another size, switch to a plugin that supports it (e.g. Phaze-A multiples of 16)

Example fix

# before (dlight_config.json)
"output_size": 160

# after
"output_size": 256

output_size = 128  # from your config
if output_size not in (128, 256, 384):
    raise SystemExit(f"Dlight output_size must be 128, 256 or 384, got {output_size}")

def is_valid_dlight_output_size(size: int) -> bool:
    """True if size is one of Dlight's supported output sizes."""
    return size in (128, 256, 384)

When it happens

Trigger: Starting training (or convert with Dlight) with output_size set to anything other than 128/256/384 in dlight config, e.g. 64 or 160. The dict {128:2, 256:4, 384:6}[cfg.output_size()] raises KeyError.

Common situations: User sets an arbitrary output size in the GUI/config assuming all sizes are supported, or copies a config from another plugin that permits different sizes.

Related errors

• 'Learn Mask' has been selected but you have not chosen a Mas
• You have requested to train with the '{self.name}' plugin, b
• Phaze-A output shape must be a multiple of 16
• No display detected. GUI mode has been disabled.
• Config file does not exist at: {ini_path}

AI-assisted analysis of deepfakes/faceswap@f530cb7508 (2026-08-15). Data as JSON: /api/errors/f0084b1da564d5b4. Report an issue: GitHub.