Skip to content

module adabmDCA.dca

function get_seqid

get_seqid(s1: Tensor, s2: Optional[Tensor] = None) → Tensor

Returns a tensor containing the sequence identities between two sets of one-hot encoded sequences. - If s2 is provided, computes the sequence identity between the corresponding sequences in s1 and s2. - If s2 is a single sequence (L, q), it computes the sequence identities between the dataset s1 and s2. - If s2 is none, computes the sequence identity between s1 and a permutation of s1.

Args:

  • s1 (torch.Tensor): One-hot encoded sequence dataset 1 of shape (batch_size, L, q) or (L, q).
  • s2 (Optional[torch.Tensor]): One-hot encoded sequence dataset 2 of shape (batch_size, L, q) or (L, q). Defaults to None.

Returns:

  • torch.Tensor: Tensor of sequence identities.

function get_seqid_stats

get_seqid_stats(s1: Tensor, s2: Optional[Tensor] = None) → Tuple[Tensor, Tensor]
  • If s2 is provided, computes the mean and the standard deviation of the mean sequence identity between two sets of one-hot encoded sequences.
  • If s2 is a single sequence (L, q), it computes the mean and the standard deviation of the mean sequence identity between the dataset s1 and s2.
  • If s2 is none, computes the mean and the standard deviation of the mean of the sequence identity between s1 and a permutation of s1.

Args:

  • s1 (torch.Tensor): One-hot encoded sequence dataset 1 of shape (batch_size, L, q) or (L, q).
  • s2 (Optional[torch.Tensor]): One-hot encoded sequence dataset 2 of shape (batch_size, L, q) or (L, q). Defaults to None.

Returns: Tuple[torch.Tensor, torch.Tensor]: (torch.Tensor) Mean sequence identity (torch.Tensor) Standard deviation of the mean sequence identity.

function set_zerosum_gauge

set_zerosum_gauge(params: Dict[str, Tensor]) → Dict[str, Tensor]

Sets the zero-sum gauge on the coupling matrix.

Args:

  • params (Dict[str, torch.Tensor]): Parameters of the model.

Returns:

  • Dict[str, torch.Tensor]: New dictionary with modified coupling matrix.
  • "bias": torch.Tensor of shape (L, q)
  • "coupling_matrix": torch.Tensor of shape (L, q, L, q)

function get_contact_map

get_contact_map(params: Dict[str, Tensor], tokens: str) → ndarray

Computes the contact map from the model coupling matrix.

Args:

  • params (Dict[str, torch.Tensor]): Model parameters. Should contain: - "coupling_matrix": torch.Tensor of shape (L, q, L, q) - "bias": torch.Tensor of shape (L, q)
  • tokens (str): Alphabet to be used.

Returns:

  • np.ndarray: Contact map.

function get_mf_contact_map

get_mf_contact_map(
    data: Tensor,
    tokens: str,
    weights: Optional[Tensor] = None
) → ndarray

Computes the contact map using mean-field approximation from the data.

Args:

  • data (torch.Tensor): Input one-hot data tensor.
  • tokens (str): Alphabet to be used.
  • weights (Optional[torch.Tensor]): Weights for the data points. Defaults to None.

Returns:

  • np.ndarray: Contact map.

This file was automatically generated via lazydocs.