babysor/MockingBird · error · NotImplementedError

Support only linear or conv1d.

Error message

Support only linear or conv1d.

What it means

Raised when positionwise_layer_type in the Conformer encoder is neither 'linear' (PositionwiseFeedForward) nor 'conv1d' (ConvolutionalPositionwiseFeedForward).

Source

Thrown at models/ppg_extractor/encoder/conformer_encoder.py:169

            )
        elif positionwise_layer_type == "conv1d":
            positionwise_layer = MultiLayeredConv1d
            positionwise_layer_args = (
                attention_dim,
                linear_units,
                positionwise_conv_kernel_size,
                dropout_rate,
            )
        elif positionwise_layer_type == "conv1d-linear":
            positionwise_layer = Conv1dLinear
            positionwise_layer_args = (
                attention_dim,
                linear_units,
                positionwise_conv_kernel_size,
                dropout_rate,
            )
        else:
            raise NotImplementedError("Support only linear or conv1d.")

        if selfattention_layer_type == "selfattn":
            logging.info("encoder self-attention layer type = self-attention")
            encoder_selfattn_layer = MultiHeadedAttention
            encoder_selfattn_layer_args = (
                attention_heads,
                attention_dim,
                attention_dropout_rate,
            )
        elif selfattention_layer_type == "rel_selfattn":
            assert pos_enc_layer_type == "rel_pos"
            encoder_selfattn_layer = RelPositionMultiHeadedAttention
            encoder_selfattn_layer_args = (
                attention_heads,
                attention_dim,
                attention_dropout_rate,
            )
        else:

View on GitHub (pinned to 28dc5e14f1)

Solutions

1. Set positionwise_layer_type to 'linear' or 'conv1d' (conformer default is usually 'conv1d')
2. Verify the YAML key spelling (positionwise_layer_type, not positionwise_type)

Example fix

# before
positionwise_layer_type: conv

# after
positionwise_layer_type: conv1d

assert positionwise_layer_type in ('linear', 'conv1d'), positionwise_layer_type

When it happens

Trigger: Constructing the conformer encoder with positionwise_layer_type set to anything other than 'linear' or 'conv1d'.

Common situations: Config typos like 'conv1' or 'ffn'; copying configs from other ESPnet forks with extra positionwise types.

Understand the failure class

Background: Invalid enum value errors: "Unknown type", "Invalid scope", "must be one of" — when a string is not on the library's allowed list — this error's family across 23 libraries.

Related errors

• unknown pos_enc_layer:
• unknown input_layer:
• unknown encoder_attn_layer:
• Support only linear or conv1d.
• unknown pos_enc_layer:

AI-assisted analysis of babysor/MockingBird@28dc5e14f1 (2026-08-27). Data as JSON: /api/errors/2ca7ac65c5052e2c. Report an issue: GitHub.