flexs.baselines.explorers.random

Defines the Random explorer class.

class flexs.baselines.explorers.random.Random(model, rounds, starting_sequence, sequences_batch_size, model_queries_per_batch, alphabet, mu=1, elitist=False, seed=None, log_file=None)

Bases: flexs.explorer.Explorer

A simple random explorer.

Chooses a random previously measured sequence and mutates it.

A good baseline to compare other search strategies against.

Since random search is not data-driven, the model is only used to score sequences, but not to guide the search strategy.

propose_sequences(measured_sequences)

Propose top sequences_batch_size sequences for evaluation.

Return type

Tuple[ndarray, ndarray]