base

class thefittest.base.EphemeralNode(generator: Callable)

class thefittest.base.FunctionalNode(value: Any, sign: str | None = None)

compile_torch(device: 'torch.device' | str | None = None) → Net

copy() → Net

cpu() → Net

cuda() → Net

ensure_compiled() → None

forward(X: torch.Tensor, weights: 'torch.Tensor' | None = None, keep_weight_dim: bool = False, autocast_input: bool = True) → torch.Tensor

get_graph() → Dict

plot(ax=None) → None

signature() → str

to(device: str | 'torch.device', dtype: 'torch.dtype' | None = None) → Net

class thefittest.base.TerminalNode(value: Any, name: str)

class thefittest.base.Tree(nodes: List[FunctionalNode | TerminalNode | EphemeralNode], n_args: List[int] | ndarray[tuple[Any, ...], dtype[int64]] | None = None)

concat(index: int, other_tree: Tree) → Tree

copy() → Tree

classmethod full_growing_method(uniset: UniversalSet, max_level: int) → Tree

get_args_id(index: int) → ndarray[tuple[Any, ...], dtype[int64]]

get_common_region(other_trees: List | ndarray[tuple[Any, ...], dtype[_ScalarT]]) → Tuple

get_graph(keep_id: bool = False) → Dict

get_levels(index: int) → ndarray[tuple[Any, ...], dtype[int64]]

get_max_level() → int64

classmethod growing_method(uniset: UniversalSet, max_level: int) → Tree

plot(ax: Any = None) → None

classmethod random_tree(uniset: UniversalSet, max_level: int) → Tree

set_terminals(**kwargs: Any) → Tree

signature() → str

subtree(index: int) → Tree

subtree_id(index: int) → Tuple[int, int]

class thefittest.base.UniversalSet(functional_set: Tuple[FunctionalNode, ...], terminal_set: Tuple[TerminalNode | EphemeralNode, ...])

thefittest.base.init_symbolic_regression_uniset(X: ndarray[tuple[Any, ...], dtype[float64]], functional_set_names: Tuple[str, ...] = ('cos', 'sin', 'add', 'sub', 'mul', 'div'), ephemeral_node_generators: Tuple[Callable, ...] | None = None)