mofaflex.ModelOptions

class mofaflex.ModelOptions(*, n_factors=0, weight_prior='Normal', factor_prior='Normal', likelihoods=None, nonnegative_weights=False, guiding_vars_likelihoods='Normal', guiding_vars_scales=1.0, nonnegative_factors=False, annotation_confidence=0.99, init_factors='random', init_scale=0.1)

Options for the model.

Attributes table

annotation_confidence	Confidence in the provided feature annotation.
factor_prior	Factor priors for each group (if dict) or for all groups (if str).
guiding_vars_likelihoods	Likelihood for each guiding variable (if dict) or for all guiding variables (if str).
guiding_vars_scales	Scale for the likelihood of each guiding variable, to put more or less emphasis on them during training.
init_factors	Initialization method for factors.
init_scale	Initialization scale of Normal distribution for factors.
likelihoods	Data likelihoods for each view (if dict) or for all views (if str).
n_factors	Number of latent factors.
nonnegative_factors	Non-negativity constraints for factors for each group (if dict) or for all groups (if bool).
nonnegative_weights	Non-negativity constraints for weights for each view (if dict) or for all views (if bool).
weight_prior	Weight priors for each view (if dict) or for all views (if str).

Methods table

Attributes

ModelOptions.annotation_confidence: float = 0.99

Confidence in the provided feature annotation. Must be between 0 and 1. Smaller values make the model more likely to add features to the annotated pathways during training, while larger values encourage the model to more closely adhere to the provided annotations.

ModelOptions.factor_prior: Mapping[str, Literal['GP', 'Horseshoe', 'Normal', 'Laplace', 'SnS']] | Literal['GP', 'Horseshoe', 'Normal', 'Laplace', 'SnS'] = 'Normal'

Factor priors for each group (if dict) or for all groups (if str).

ModelOptions.guiding_vars_likelihoods: Mapping[str, str] | Literal['Normal', 'Categorical', 'Bernoulli'] | None = 'Normal'

Likelihood for each guiding variable (if dict) or for all guiding variables (if str).

ModelOptions.guiding_vars_scales: Mapping[str, float] | float = 1.0

Scale for the likelihood of each guiding variable, to put more or less emphasis on them during training.

ModelOptions.init_factors: float | Literal['random', 'orthogonal', 'pca'] = 'random'

Initialization method for factors.

ModelOptions.init_scale: float = 0.1

Initialization scale of Normal distribution for factors.

ModelOptions.likelihoods: Mapping[str, Literal['Bernoulli', 'NegativeBinomial', 'Normal']] | Literal['Bernoulli', 'NegativeBinomial', 'Normal'] | None = None

Data likelihoods for each view (if dict) or for all views (if str). Inferred automatically if None.

ModelOptions.n_factors: int = 0

Number of latent factors.

ModelOptions.nonnegative_factors: Mapping[str, bool] | bool = False

Non-negativity constraints for factors for each group (if dict) or for all groups (if bool).

ModelOptions.nonnegative_weights: Mapping[str, bool] | bool = False

Non-negativity constraints for weights for each view (if dict) or for all views (if bool).

ModelOptions.weight_prior: Mapping[str, Literal['Horseshoe', 'Normal', 'Laplace', 'SnS']] | Literal['Horseshoe', 'Normal', 'Laplace', 'SnS'] = 'Normal'

Weight priors for each view (if dict) or for all views (if str).

Methods