IsingModel

class abess.graph.IsingModel

Sparse Ising model estimation via SLIDE (Sparse pairwise Logistic regression for Ising model via best-subset sElection).

Fits one LogisticRegression per node (node-wise logistic regression), then symmetrizes the resulting coefficient matrix.

Parameters

• max_support_size (int, array-like of shape (p,), or None) -- Maximum support size for each node's logistic regression. If None, defaults to min(p-2, 100) for each node.

• tune_type (str, optional) -- Model selection criterion: "gic", "aic", "bic", "ebic", or "cv". Default: "gic".

• ic_coef (float, optional) -- Coefficient for information criterion. Default: 1.0.

• cv (int, optional) -- Number of cross-validation folds (used when tune_type="cv"). Default: 5.

• graph_threshold (float, optional) -- Entries with |theta| <= graph_threshold are zeroed out. Default: 0.0.

• thread (int, optional) -- Number of threads for LogisticRegression. Default: 1.

• primary_model_fit_max_iter (int, optional) -- Max iterations for inner logistic solver. Default: 100.

• primary_model_fit_epsilon (float, optional) -- Convergence tolerance for inner logistic solver. Default: 1e-6.

__init__(max_support_size=None, tune_type='gic', ic_coef=1.0, cv=5, graph_threshold=0.0, thread=1, primary_model_fit_max_iter=100, primary_model_fit_epsilon=1e-06)

fit(X, y=None, sample_weight=None)

Fit the sparse Ising model via SLIDE.

Parameters

• X (array-like of shape (n_samples, p_features)) -- Binary data with values in {-1, 1}.

• y (ignored) --

• sample_weight (array-like of shape (n_samples,), optional) -- Sample weights.

Return type

self

score(X, y=None, sample_weight=None)

Compute the pseudo-log-likelihood of X under the fitted Ising model.

Parameters

• X (array-like of shape (n_samples, p_features)) --

• y (ignored) --

• sample_weight (ignored) --

Returns

Mean pseudo-log-likelihood (higher is better).

Return type

float

get_params(deep=True)

Get parameters for this estimator.

Parameters

deep (bool, default=True) -- If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns

params -- Parameter names mapped to their values.

Return type

dict

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it's possible to update each component of a nested object.

Parameters

**params (dict) -- Estimator parameters.

Returns

self -- Estimator instance.

Return type

estimator instance

__new__(*args, **kwargs)