hankcs/HanLP · error · ValueError

self.model.config.pad_token_id has to be defined.

Error message

self.model.config.pad_token_id has to be defined.

What it means

Feedforward accepts dropout as a scalar (broadcast to all layers) or a per-layer list; if you pass a list its length must equal num_layers, otherwise the constructor raises ValueError.

Source

Thrown at hanlp/components/amr/amrbart/model_interface/modeling_bart.py:84

_QA_EXPECTED_OUTPUT = "' nice puppet'"


BART_PRETRAINED_MODEL_ARCHIVE_LIST = [
    "facebook/bart-large",
    # see all BART models at https://huggingface.co/models?filter=bart
]


def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decoder_start_token_id: int):
    """
    Shift input ids one token to the right.
    """
    shifted_input_ids = input_ids.new_zeros(input_ids.shape)
    shifted_input_ids[:, 1:] = input_ids[:, :-1].clone()
    shifted_input_ids[:, 0] = decoder_start_token_id

    if pad_token_id is None:
        raise ValueError("self.model.config.pad_token_id has to be defined.")
    # replace possible -100 values in labels by `pad_token_id`
    shifted_input_ids.masked_fill_(shifted_input_ids == -100, pad_token_id)

    return shifted_input_ids


def _make_causal_mask(input_ids_shape: torch.Size, dtype: torch.dtype, past_key_values_length: int = 0):
    """
    Make causal mask used for bi-directional self-attention.
    """
    bsz, tgt_len = input_ids_shape
    mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min))
    mask_cond = torch.arange(mask.size(-1))
    mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
    mask = mask.to(dtype)

    if past_key_values_length > 0:
        mask = torch.cat([torch.zeros(tgt_len, past_key_values_length, dtype=dtype), mask], dim=-1)

View on GitHub (pinned to ddb1299bdd)

Solutions

1. Pass a scalar dropout (e.g. dropout=0.5) to apply uniformly
2. Set len(dropout) == num_layers
3. Validate generated sweep configs before model construction

Example fix

# before
Feedforward(input_dim=300, num_layers=3, hidden_dims=[128]*3, dropout=[0.2, 0.5])
# after
Feedforward(input_dim=300, num_layers=3, hidden_dims=[128]*3, dropout=0.5)

assert isinstance(dropout, (int, float)) or len(dropout) == num_layers

try:
    ff = Feedforward(..., dropout=dropout)
except ValueError:
    ff = Feedforward(..., dropout=0.5)

When it happens

Trigger: Passing dropout=[0.2, 0.5] with num_layers=3, or dropout=0.5 as a string/list with the wrong length via config.

Common situations: Hyperparameter sweeps generating per-layer dropout lists of the wrong length; editing configs and changing layer counts; JSON configs where a scalar was replaced by a list.

Understand the failure class

Background: Config validation failed: what "invalid value for {key}" and settings-rejection errors mean across 19 open-source libraries — this error's family across 19 libraries.

Related errors

• Unsupported argument type: {item}
• Unrecognized mapper type {mapper}
• embed_dim must be divisible by num_heads (got `embed_dim`: {
• You cannot specify both input_ids and inputs_embeds at the s
• Unsupported dim: {x.dim()}. Only 2d (T,C) or 3d (B,T,C) is s

AI-assisted analysis of hankcs/HanLP@ddb1299bdd (2026-08-27). Data as JSON: /api/errors/14b01dbe0462b780. Report an issue: GitHub.