U ?h@sbddlZddlmZddlmZddlmZdgZddddd dZ d d Z dd d Z ddZ dS)N)contingency_table)check_shape_equalityvariation_of_informationtable ignore_labelscCs*t||||d\}}t||gS)uReturn symmetric conditional entropies associated with the VI. [1]_ The variation of information is defined as VI(X,Y) = H(X|Y) + H(Y|X). If X is the ground-truth segmentation, then H(X|Y) can be interpreted as the amount of under-segmentation and H(Y|X) as the amount of over-segmentation. In other words, a perfect over-segmentation will have H(X|Y)=0 and a perfect under-segmentation will have H(Y|X)=0. Parameters ---------- image0, image1 : ndarray of int Label images / segmentations, must have same shape. table : scipy.sparse array in csr format, optional A contingency table built with skimage.evaluate.contingency_table. If None, it will be computed with skimage.evaluate.contingency_table. If given, the entropies will be computed from this table and any images will be ignored. ignore_labels : sequence of int, optional Labels to ignore. Any part of the true image labeled with any of these values will not be counted in the score. Returns ------- vi : ndarray of float, shape (2,) The conditional entropies of image1|image0 and image0|image1. References ---------- .. [1] Marina Meilă (2007), Comparing clusterings—an information based distance, Journal of Multivariate Analysis, Volume 98, Issue 5, Pages 873-895, ISSN 0047-259X, :DOI:`10.1016/j.jmva.2006.11.013`. r) _vi_tablesnparraysum)Zimage0Zimage1r r Zh0g1Zh1g0rr[/var/www/html/venv/lib/python3.8/site-packages/skimage/metrics/_variation_of_information.pyr s" cCsX|}t|tjs t|tjr(|j}n t|}|}||t ||9<|S)zCompute x * log_2(x). We define 0 * log_2(0) = 0 Parameters ---------- x : ndarray or scipy.sparse.csc_matrix or csr_matrix The input array. Returns ------- y : same type as x Result of x * log_2(x). ) copy isinstancesparseZ csc_matrixZ csr_matrixdatar asarraynonzerolog2)xyznzrrr_xlogx1s rc Cst|||dkr$t|||dd}n|}t|jdd}t|jdd}tt|}tt|}| t||jdd} t||jdd |} t t tj | | gS)aCompute probability tables used for calculating VI. Parameters ---------- im_true, im_test : ndarray of int Input label images, any dimensionality. table : csr matrix, optional Pre-computed contingency table. ignore_labels : sequence of int, optional Labels to ignore when computing scores. Returns ------- hxgy, hygx : ndarray of float Per-segment conditional entropies of ``im_true`` given ``im_test`` and vice-versa. NT)r normalizer)Zaxisr) rrr ZravelrrZdiags_invert_nonzerorlistmapr) Zim_trueZim_testr r ZpxyZpxpyZpx_invZpy_invZhygxZhxgyrrrr Js  r cCs&|}t|}d||||<|S)a Compute the inverse of the non-zero elements of arr, not changing 0. Parameters ---------- arr : ndarray Returns ------- arr_inv : ndarray Array containing the inverse of the non-zero elements of arr, and zero elsewhere. r)rr r)ZarrZarr_invrrrrrxs  r)NN)Nr) numpyr Z scipy.sparserZ_contingency_tablerZ _shared.utilsr__all__rrr rrrrrs    ( .