hankcs/HanLP · error · ValueError

If no `decoder_input_ids` or `decoder_inputs_embeds` are pas

Error message

If no `decoder_input_ids` or `decoder_inputs_embeds` are passed, `input_ids` cannot be `None`. Please pass either `input_ids` or `decoder_input_ids` or `decoder_inputs_embeds`.

What it means

SmatchEval for AMR evaluation downloads version-specific official utility scripts; only AMR corpus versions '1.0', '2.0' and '3.0' are supported. Any other amr_version string raises ValueError from get_amr_utils, which is called during __init__ and post_process.

Source

Thrown at hanlp/components/amr/amrbart/model_interface/modeling_bart.py:1204

        decoder_attention_mask: Optional[torch.LongTensor] = None,
        head_mask: Optional[torch.Tensor] = None,
        decoder_head_mask: Optional[torch.Tensor] = None,
        cross_attn_head_mask: Optional[torch.Tensor] = None,
        encoder_outputs: Optional[List[torch.FloatTensor]] = None,
        past_key_values: Optional[List[torch.FloatTensor]] = None,
        inputs_embeds: Optional[torch.FloatTensor] = None,
        decoder_inputs_embeds: Optional[torch.FloatTensor] = None,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, Seq2SeqModelOutput]:

        # different to other models, Bart automatically creates decoder_input_ids from
        # input_ids if no decoder_input_ids are provided
        if decoder_input_ids is None and decoder_inputs_embeds is None:
            if input_ids is None:
                raise ValueError(
                    "If no `decoder_input_ids` or `decoder_inputs_embeds` are "
                    "passed, `input_ids` cannot be `None`. Please pass either "
                    "`input_ids` or `decoder_input_ids` or `decoder_inputs_embeds`."
                )

            decoder_input_ids = shift_tokens_right(
                input_ids, self.config.pad_token_id, self.config.decoder_start_token_id
            )

        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
        output_hidden_states = (
            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
        )
        use_cache = use_cache if use_cache is not None else self.config.use_cache
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        if encoder_outputs is None:
            encoder_outputs = self.encoder(

View on GitHub (pinned to ddb1299bdd)

Solutions

1. Use one of '1.0', '2.0', '3.0' exactly
2. Map your corpus to the closest supported version (e.g. AMR 2.0 utilities for AMR 2.5 data, if structurally compatible)
3. Upgrade HanLP to a version supporting your AMR version

Example fix

# before
eval = SmatchEval(amr_version='2.5')
# after
eval = SmatchEval(amr_version='2.0')

assert amr_version in ('1.0', '2.0', '3.0'), f'unsupported AMR version {amr_version!r}'

def supported_amr_version(v: str) -> bool:
    return v in ('1.0', '2.0', '3.0')

try:
    ev = SmatchEval(amr_version=amr_version)
except ValueError:
    amr_version = '2.0'
    ev = SmatchEval(amr_version=amr_version)

When it happens

Trigger: Constructing SmatchEval(amr_version='2.5') or any string other than 1.0/2.0/3.0, including forms like 'v2' or 'AMR2.0'.

Common situations: New AMR corpus releases not yet supported by this HanLP version; typo'd or unnormalized version strings in configs; passing the corpus name instead of its version.

Related errors

• Failed to parse results from smatch: {line}
• Unsupported argument length: {item}
• error
• output ({}) must be of type bool or str
• Call fit or load before evaluate.

AI-assisted analysis of hankcs/HanLP@ddb1299bdd (2026-08-27). Data as JSON: /api/errors/83172f13b1c0d4d6. Report an issue: GitHub.