NumpyActiveLearningPool

class probly.evaluation.active_learning.pool.array.NumpyActiveLearningPool(x_labeled: ndarray, y_labeled: ndarray, x_unlabeled: ndarray, y_unlabeled: ndarray, x_test: ndarray, y_test: ndarray)[source]

Bases: object

Active learning pool backed by NumPy arrays.

Variables:

  • x_labeled (numpy.ndarray) – Feature matrix for the currently labeled training samples.

  • y_labeled (numpy.ndarray) – Labels for the currently labeled training samples.

  • x_unlabeled (numpy.ndarray) – Feature matrix for the unlabeled pool.

  • y_unlabeled (numpy.ndarray) – Ground-truth labels for the unlabeled pool (revealed only on query).

  • x_test (numpy.ndarray) – Feature matrix for the held-out test set.

  • y_test (numpy.ndarray) – Labels for the held-out test set.

property n_labeled: int

Number of currently labeled training samples.

property n_unlabeled: int

Number of samples remaining in the unlabeled pool.

x_labeled: ndarray
x_test: ndarray
x_unlabeled: ndarray
y_labeled: ndarray
y_test: ndarray
y_unlabeled: ndarray

Examples using probly.evaluation.active_learning.pool.array.NumpyActiveLearningPool

Active Learning with sklearn - Margin vs Random

Active Learning with sklearn - Margin vs Random