Importance-Weighted Classifier¶
-
class
libtlda.iw.ImportanceWeightedClassifier(loss_function='logistic', l2_regularization=None, weight_estimator='lr', smoothing=True, clip_max_value=-1, kernel_type='rbf', bandwidth=1)¶ Class of importance-weighted classifiers.
Methods contain different importance-weight estimators and different loss functions.
Examples
>>>> X = np.random.randn(10, 2)>>>> y = np.vstack((-np.ones((5,)), np.ones((5,))))>>>> Z = np.random.randn(10, 2)>>>> clf = ImportanceWeightedClassifier()>>>> clf.fit(X, y, Z)>>>> u_pred = clf.predict(Z)Methods
fit(X, y, Z)Fit/train an importance-weighted classifier. get_params()Get classifier parameters. get_weights()Get estimated importance weights. is_trained()Check whether classifier is trained. iwe_kernel_densities(X, Z)Estimate importance weights based on kernel density estimation. iwe_kernel_mean_matching(X, Z)Estimate importance weights based on kernel mean matching. iwe_logistic_discrimination(X, Z)Estimate importance weights based on logistic regression. iwe_nearest_neighbours(X, Z)Estimate importance weights based on nearest-neighbours. iwe_ratio_gaussians(X, Z)Estimate importance weights based on a ratio of Gaussian distributions. predict(Z)Make predictions on new dataset. predict_proba(Z)Compute posterior probabilities on new dataset. -
fit(X, y, Z)¶ Fit/train an importance-weighted classifier.
Parameters: - X : array
source data (N samples by D features)
- y : array
source labels (N samples by 1)
- Z : array
target data (M samples by D features)
Returns: - None
-
get_params()¶ Get classifier parameters.
-
get_weights()¶ Get estimated importance weights.
-
is_trained()¶ Check whether classifier is trained.
-
iwe_kernel_densities(X, Z)¶ Estimate importance weights based on kernel density estimation.
Parameters: - X : array
source data (N samples by D features)
- Z : array
target data (M samples by D features)
Returns: - array
importance weights (N samples by 1)
-
iwe_kernel_mean_matching(X, Z)¶ Estimate importance weights based on kernel mean matching.
Parameters: - X : array
source data (N samples by D features)
- Z : array
target data (M samples by D features)
Returns: - iw : array
importance weights (N samples by 1)
-
iwe_logistic_discrimination(X, Z)¶ Estimate importance weights based on logistic regression.
Parameters: - X : array
source data (N samples by D features)
- Z : array
target data (M samples by D features)
Returns: - array
importance weights (N samples by 1)
-
iwe_nearest_neighbours(X, Z)¶ Estimate importance weights based on nearest-neighbours.
Parameters: - X : array
source data (N samples by D features)
- Z : array
target data (M samples by D features)
Returns: - iw : array
importance weights (N samples by 1)
-
iwe_ratio_gaussians(X, Z)¶ Estimate importance weights based on a ratio of Gaussian distributions.
Parameters: - X : array
source data (N samples by D features)
- Z : array
target data (M samples by D features)
Returns: - iw : array
importance weights (N samples by 1)
-
predict(Z)¶ Make predictions on new dataset.
Parameters: - Z : array
new data set (M samples by D features)
Returns: - preds : array
label predictions (M samples by 1)
-
predict_proba(Z)¶ Compute posterior probabilities on new dataset.
Parameters: - Z : array
new data set (M samples by D features)
Returns: - probs : array
label predictions (M samples by K)
-