initialize

Sample sequences from a structure using the ProteinMPNN model.

prxteinmpnn.sampling.initialize

prxteinmpnn.sampling.initialize.sampling_encode(encoder, decoding_order_fn)

Create a function to run a single pass through the encoder.

Return type:

Callable[[Unpack[tuple[Union[Key[Array, ''], UInt32[Array, '2']], PyTree[str, 'P'], Float[Array, 'num_residues num_atoms 3'], Int[Array, 'num_residues num_atoms'], Int[Array, 'num_residues'], Int[Array, 'num_residues'], Bool[Array, 'num_residues num_residues'] | None, int, Float[Array, 'n'] | None]]], tuple[Int[Array, 'num_atoms num_features'], Float[Array, 'num_atoms num_neighbors num_features'], Int[Array, 'num_atoms num_neighbors'], Int[Array, 'num_residues num_atoms'], Bool[Array, 'num_residues num_residues'], Union[Key[Array, ''], UInt32[Array, '2']]]]

Parameters:

• encoder (Callable[[...], tuple[Int[Array, 'num_atoms num_features'], Float[Array, 'num_atoms num_neighbors num_features']]])

• decoding_order_fn (Callable[[Unpack[tuple[Key[Array, ''] | UInt32[Array, '2'], Int]]], tuple[Int[Array, 'num_residues'], Key[Array, ''] | UInt32[Array, '2']]] | None)