modeling_outputs

multimolecule.models.modeling_outputs

SequencePredictorOutput dataclass

Bases: ModelOutput

Base class for outputs of sentence classification & regression models.

Parameters:

Name	Type	Description	Default
loss	FloatTensor | None	torch.FloatTensor of shape (1,). Optional, returned when labels is provided	None
logits	FloatTensor	torch.FloatTensor of shape (batch_size, sequence_length, config.num_labels) Prediction outputs.	None
hidden_states	Tuple[FloatTensor, ...] | None	Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size). Optional, returned when output_hidden_states=True is passed or when `config.output_hidden_states=True Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.	None
attentions	Tuple[FloatTensor, ...] | None	Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length). Optional, eturned when output_attentions=True is passed or when config.output_attentions=True Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.	None

Source code in multimolecule/models/modeling_outputs.py

@dataclass
class SequencePredictorOutput(ModelOutput):
    """
    Base class for outputs of sentence classification & regression models.

    Args:
        loss:
            `torch.FloatTensor` of shape `(1,)`.

            Optional, returned when `labels` is provided
        logits:
            `torch.FloatTensor` of shape `(batch_size, sequence_length, config.num_labels)`

            Prediction outputs.
        hidden_states:
            Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, +
            one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`.

            Optional, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True

            Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
        attentions:
            Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length,
            sequence_length)`.

            Optional, eturned when `output_attentions=True` is passed or when `config.output_attentions=True`

            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
            heads.
    """

    loss: torch.FloatTensor | None = None
    logits: torch.FloatTensor = None
    hidden_states: Tuple[torch.FloatTensor, ...] | None = None
    attentions: Tuple[torch.FloatTensor, ...] | None = None

TokenPredictorOutput dataclass

Bases: ModelOutput

Base class for outputs of token classification & regression models.

Parameters:

Name	Type	Description	Default
loss	FloatTensor | None	torch.FloatTensor of shape (1,). Optional, returned when labels is provided	None
logits	FloatTensor	torch.FloatTensor of shape (batch_size, sequence_length, config.num_labels) Prediction outputs.	None
hidden_states	Tuple[FloatTensor, ...] | None	Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size). Optional, returned when output_hidden_states=True is passed or when `config.output_hidden_states=True Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.	None
attentions	Tuple[FloatTensor, ...] | None	Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length). Optional, eturned when output_attentions=True is passed or when config.output_attentions=True Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.	None

Source code in multimolecule/models/modeling_outputs.py

@dataclass
class TokenPredictorOutput(ModelOutput):
    """
    Base class for outputs of token classification & regression models.

    Args:
        loss:
            `torch.FloatTensor` of shape `(1,)`.

            Optional, returned when `labels` is provided
        logits:
            `torch.FloatTensor` of shape `(batch_size, sequence_length, config.num_labels)`

            Prediction outputs.
        hidden_states:
            Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, +
            one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`.

            Optional, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True

            Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
        attentions:
            Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length,
            sequence_length)`.

            Optional, eturned when `output_attentions=True` is passed or when `config.output_attentions=True`

            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
            heads.
    """

    loss: torch.FloatTensor | None = None
    logits: torch.FloatTensor = None
    hidden_states: Tuple[torch.FloatTensor, ...] | None = None
    attentions: Tuple[torch.FloatTensor, ...] | None = None

ContactPredictorOutput dataclass

Bases: ModelOutput

Base class for outputs of contact classification & regression models.

Parameters:

Name	Type	Description	Default
loss	FloatTensor | None	torch.FloatTensor of shape (1,). Optional, returned when labels is provided	None
logits	FloatTensor	torch.FloatTensor of shape (batch_size, sequence_length, config.num_labels) Prediction outputs.	None
hidden_states	Tuple[FloatTensor, ...] | None	Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size). Optional, returned when output_hidden_states=True is passed or when `config.output_hidden_states=True Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.	None
attentions	Tuple[FloatTensor, ...] | None	Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length). Optional, eturned when output_attentions=True is passed or when config.output_attentions=True Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.	None

Source code in multimolecule/models/modeling_outputs.py

@dataclass
class ContactPredictorOutput(ModelOutput):
    """
    Base class for outputs of contact classification & regression models.

    Args:
        loss:
            `torch.FloatTensor` of shape `(1,)`.

            Optional, returned when `labels` is provided
        logits:
            `torch.FloatTensor` of shape `(batch_size, sequence_length, config.num_labels)`

            Prediction outputs.
        hidden_states:
            Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, +
            one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`.

            Optional, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True

            Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
        attentions:
            Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length,
            sequence_length)`.

            Optional, eturned when `output_attentions=True` is passed or when `config.output_attentions=True`

            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
            heads.
    """

    loss: torch.FloatTensor | None = None
    logits: torch.FloatTensor = None
    hidden_states: Tuple[torch.FloatTensor, ...] | None = None
    attentions: Tuple[torch.FloatTensor, ...] | None = None