h`SSKrSSKJr SSKJr SSKJr S/rS SSS.S jjr S r SS jr S r g)N)contingency_table)check_shape_equalityvariation_of_informationtable ignore_labelsc[XX#S9upE[R"UR5UR5/5$)u.Return symmetric conditional entropies associated with the VI. [1]_ The variation of information is defined as VI(X,Y) = H(X|Y) + H(Y|X). If X is the ground-truth segmentation, then H(X|Y) can be interpreted as the amount of under-segmentation and H(Y|X) as the amount of over-segmentation. In other words, a perfect over-segmentation will have H(X|Y)=0 and a perfect under-segmentation will have H(Y|X)=0. Parameters ---------- image0, image1 : ndarray of int Label images / segmentations, must have same shape. table : scipy.sparse array in csr format, optional A contingency table built with skimage.evaluate.contingency_table. If None, it will be computed with skimage.evaluate.contingency_table. If given, the entropies will be computed from this table and any images will be ignored. ignore_labels : sequence of int, optional Labels to ignore. Any part of the true image labeled with any of these values will not be counted in the score. Returns ------- vi : ndarray of float, shape (2,) The conditional entropies of image1|image0 and image0|image1. References ---------- .. [1] Marina Meilă (2007), Comparing clusterings—an information based distance, Journal of Multivariate Analysis, Volume 98, Issue 5, Pages 873-895, ISSN 0047-259X, :DOI:`10.1016/j.jmva.2006.11.013`. r ) _vi_tablesnparraysum)image0image1r r h0g1h1g0s c/var/www/fran/franai/venv/lib/python3.13/site-packages/skimage/metrics/_variation_of_information.pyrr s3BF%UJD 88TXXZ, --c&UR5n[R"U5(aURS;a URnO[ R "U5nUR5nX#==[ R"X#5-ss'U$)zCompute x * log_2(x). We define 0 * log_2(0) = 0 Parameters ---------- x : ndarray or scipy.sparse.csc_array or scipy.sparse.csr_array The input array. Returns ------- y : same type as x Result of x * log_2(x). )csccsr) copysparseissparseformatdatarasarraynonzerolog2)xyznzs r_xlogxr&/se A qahh.8 FF JJqM BERWWQU^E Hrc^[X5 Uc [XUSS9nOUn[R"UR SS95n[R"UR SS95n[ R "[U5S4URUR4S9n[ R "[U5S4URUR4S9nU*[Xt-5R SS9-n [XH-5R SS9*U-n [[[RX/55$)aCompute probability tables used for calculating VI. Parameters ---------- im_true, im_test : ndarray of int Input label images, any dimensionality. table : csr_array, 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. T)r normalizer)axisr)shape) rrrravelrr dia_array_invert_nonzerosizer&listmapr) im_trueim_testr r pxypxpypx_invpy_invhygxhxgys rr r Hs $* } MT   #''q'/ "B #''q'/ "B  r2A6rww>P QF   r2A6rww>P QF 3 %))q)1 1D 3< $ $! $ , ,r 1D BJJ - ..rcdUR5n[R"U5nSX- X'U$)zCompute 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)rrr )arrarr_invr%s rr-r-us.hhjG CBcg+GK Nr)NN)Nr) numpyr scipy.sparser_contingency_tabler _shared.utilsr__all__rr&r r-rrrrBs910 % &#.TV#.L 2*/Zr