torch_mist.decomposition.base
Module Contents
Classes
Attributes
- torch_mist.decomposition.base.DEFAULT_MAX_ITERATIONS = 5000
- torch_mist.decomposition.base.DEFAULT_BATCH_SIZE = 64
- class torch_mist.decomposition.base.CenterAndScale(loc: torch.Tensor | numpy.ndarray, scale: torch.Tensor | numpy.ndarray, min_scale=1e-06)
- __call__(data)
- class torch_mist.decomposition.base.DimensionalityReduction(n_dim: int, normalize_inputs: bool = True, whiten: bool = False, proj: torch.nn.Module | None = None, y_proj: torch.nn.Module | None = None, model: torch_mist.nn.Model | None = None, proj_params: Dict[str, Any] | None = None, y_proj_params: Dict[str, Any] | None = None, model_params: Dict[str, Any] | None = None)
Bases:
sklearn.base.TransformerMixin- _add_default_model_params(model_params: Dict[str, Any] | None) Dict[str, Any] | None
- _add_default_proj_params(proj_params: Dict[str, Any] | None) Dict[str, Any] | None
- _add_default_y_proj_params(y_proj_params: Dict[str, Any] | None) Dict[str, Any] | None
- _add_default_train_params(train_params: Dict[str, Any] | None) Dict[str, Any]
- abstract _instantiate_proj(x_dim: int) torch.nn.Module
- abstract _instantiate_y_proj(y_dim: int) torch.nn.Module | None
- abstract _instantiate_model(x_dim: int, y_dim: int) torch_mist.nn.Model
- _train_model(data: torch_mist.utils.data.utils.TensorDictLike, **train_params)
- _get_transformed_y_dim(Y: numpy.ndarray) int
- fit(X: numpy.ndarray | torch.Tensor, Y: numpy.ndarray | torch.Tensor, **train_params)
- _encode(X: torch.Tensor) torch.Tensor
- transform(X: numpy.ndarray) numpy.ndarray | torch.Tensor
- class torch_mist.decomposition.base.StochasticDimensionalityReduction(*args, stochastic_transform: bool = False, **kwargs)
Bases:
DimensionalityReduction- _instantiate_proj(x_dim: int) torch.nn.Module
- _get_transformed_y_dim(Y: numpy.ndarray) int
- _encode(X: torch.Tensor) torch.Tensor