`torch_mist.decomposition.ceb`

Module Contents

Classes

Attributes

`DEFAULT_MAX_ITERATIONS`
`DEFAULT_BATCH_SIZE`

torch_mist.decomposition.ceb.DEFAULT_MAX_ITERATIONS = 5000

torch_mist.decomposition.ceb.DEFAULT_BATCH_SIZE = 64

class torch_mist.decomposition.ceb.CEB(*args, beta: float = 0.01, conditional_dist_params: Dict[str, Any] | None = None, **kwargs)

Bases: torch_mist.decomposition.base.StochasticDimensionalityReduction

_add_default_conditional_dist_params(cond_params: Dict[str, Any] | None) → Dict[str, Any]

_add_default_mi_estimator_params(model_params: Dict[str, Any] | None) → Dict[str, Any]

_add_default_model_params(model_params: Dict[str, Any] | None) → Dict[str, Any]

_add_default_proj_params(proj_params: Dict[str, Any] | None) → Dict[str, Any]

_train_model(data: torch_mist.utils.data.utils.TensorDictLike, **train_params)

_instantiate_y_proj(y_dim: int) → torch.nn.Module | None

_instantiate_model(x_dim: int, y_dim: int)