3jE7SSKJrJrJr SSKrSSKJr SSKJr /SQr"SS\R5r SS\RS \ S \ S \S \ S \\R\R\\R44 SjjrSS\RS \ S \ S \S \R4 SjjrSS\RS\RS\RS \ S \R4 Sjjrg))ListOptionalTupleN)nn)_torch_svd_cast) ZCAWhiteninglinear_transformzca_mean zca_whitenc ^\rSrSrSrSS\S\S\S\S\SS 4 U4S jjjrS \ RSS4S jr SS \ RS \S\ R4Sjjr S \ RS\ R4Sjr SrU=r$)raCompute the ZCA whitening matrix transform and the mean vector and applies the transform to the data. The data torch.Tensor is flattened, and the mean :math:`\mathbf{\mu}` and covariance matrix :math:`\mathbf{\Sigma}` are computed from the flattened data :math:`\mathbf{X} \in \mathbb{R}^{N \times D}`, where :math:`N` is the sample size and :math:`D` is flattened dimensionality (e.g. for a torch.Tensor with size 5x3x2x2 :math:`N = 5` and :math:`D = 12`). The ZCA whitening transform is given by: .. math:: \mathbf{X}_{\text{zca}} = (\mathbf{X - \mu})(US^{-\frac{1}{2}}U^T)^T where :math:`U` are the eigenvectors of :math:`\Sigma` and :math:`S` contain the corresponding eigenvalues of :math:`\Sigma`. After the transform is applied, the output is reshaped to same shape. Args: dim: Determines the dimension that represents the samples axis. eps: a small number used for numerical stability. unbiased: Whether to use the biased estimate of the covariance matrix. compute_inv: Compute the inverse transform matrix. detach_transforms: Detaches gradient from the ZCA fitting. shape: - x: :math:`(D_0,...,D_{\text{dim}},...,D_N)` is a batch of N-D tensors. - x_whiten: :math:`(D_0,...,D_{\text{dim}},...,D_N)` same shape as input. .. note:: See a working example `here `__. Examples: >>> x = torch.tensor([[0,1],[1,0],[-1,0],[0,-1]], dtype = torch.float32) >>> zca = ZCAWhitening().fit(x) >>> x_whiten = zca(x) >>> zca = ZCAWhitening() >>> x_whiten = zca(x, include_fit = True) # Includes the fitting step >>> x_whiten = zca(x) # Can run now without the fitting set >>> # Enable backprop through ZCA fitting process >>> zca = ZCAWhitening(detach_transforms = False) >>> x_whiten = zca(x, include_fit = True) # Includes the fitting step Note: This implementation uses :py:meth:`~torch.svd` which yields NaNs in the backwards step if the singular values are not unique. See `here `_ for more information. References: [1] `Stanford PCA & ZCA whitening tutorial `_ dimepsunbiaseddetach_transforms compute_invreturnNcx>[TU]5 XlX lX0lX@lXPlSUlU U U g)NF)super__init__rrrrrfitted)selfrrrrr __class__s L/home/wildlama/miniconda3/lib/python3.13/site-packages/kornia/enhance/zca.pyrZCAWhitening.__init__Qs;  !2&    xc[XRURURUR5up#nX0lX lUc[R"S/5Ul OX@l UR(a]UR R5UlUR R5UlURR5Ul SUl U$)z}Fit ZCA whitening matrices to the data. Args: x: Input data. Returns: Returns a fitted ZCAWhiten object instance. rT) r rrrr mean_vectortransform_matrixtorchempty transform_invrdetachr)rrTmeanT_invs rfitZCAWhitening.fitgs"!XXt}}dhhHXHXY ! =!&aS!1D !&   ! !#//668D $($9$9$@$@$BD !!%!3!3!:!:!N>NPTPXPXYrcUR(d [S5eUR(d [S5eURc [ S5eUR R UR5*n[XRU5nU$)zmApply the inverse transform to the whitened data. Args: x: Whitened data. Returns: Original data. r,z;Did not compute inverse ZCA. Please set compute_inv to TruezTTF)F)__name__ __module__ __qualname____firstlineno____doc__intfloatboolrr!Tensorr(r/r6__static_attributes__ __classcell__)rs@rrrs2l"&! 3 33 3  3  3 33,U\\n8DU\\*5<<ELLrrinprrrreturn_inverserc [U[R5(d[S[ U535e[U[ 5(d[S[ U535e[U[ 5(d[S[ U535e[U[5(d[S[ U535e[U[ 5(d[S[ U535eUR5nU[U5:dU[U5*:a*[S[U5*S[U5S- S U35eUS :a[U5U-n[R"[R"S U5[R"US-[U55/5n[R"[R"U/5U/5R5nURU5nXQn [R"US UXQS-S -5n [[R "U 5R#55n [R$"US S S 9n U R'SU 45n UR'X45U - n U R)5R+U 5nU(aU[ U S- 5- nOU[ U 5- n[-U5unnnUR'SS5n[R."UU-5nUR+UUR)5-5nS nU(a9UR+[R0"UU-5UR)5-5nUU U4$)aCompute the ZCA whitening matrix and mean vector. The output can be used with :py:meth:`~kornia.color.linear_transform`. See :class:`~kornia.color.ZCAWhitening` for details. Args: inp: input data torch.Tensor. dim: Specifies the dimension that serves as the samples dimension. unbiased: Whether to use the unbiased estimate of the covariance matrix. eps: a small number used for numerical stability. return_inverse: Whether to return the inverse ZCA transform. Shapes: - inp: :math:`(D_0,...,D_{\text{dim}},...,D_N)` is a batch of N-D tensors. - transform_matrix: :math:`(\Pi_{d=0,d\neq \text{dim}}^N D_d, \Pi_{d=0,d\neq \text{dim}}^N D_d)` - mean_vector: :math:`(1, \Pi_{d=0,d\neq \text{dim}}^N D_d)` - inv_transform: same shape as the transform matrix Returns: A tuple containing the ZCA matrix and the mean vector. If return_inverse is set to True, then it returns the inverse ZCA matrix, otherwise it returns None. .. note:: See a working example `here `__. Examples: >>> x = torch.tensor([[0,1],[1,0],[-1,0],[0,-1]], dtype = torch.float32) >>> transform_matrix, mean_vector,_ = zca_mean(x) # Returns transformation matrix and data mean >>> x = torch.rand(3,20,2,2) >>> transform_matrix, mean_vector, inv_transform = zca_mean(x, dim = 1, return_inverse = True) >>> # transform_matrix.size() equals (12,12) and the mean vector.size equal (1,12) &Input type is not a torch.Tensor. Got eps type is not a float. Gotunbiased type is not bool. Got(Argument 'dim' must be of type int. Got z-Argument return_inverse must be of type bool 4Dimension out of range (expected to be in range of [, ], but got rNT)rkeepdim) isinstancer!rAr2typer?r@r>sizelen IndexErrorcatarangetensortolistpermuteproditemr&reshapetr3rrsqrtsqrt)rDrrrrEinp_size feat_dims new_order inp_permuteN feature_sizes num_featuresr&inp_center_flatcovUS_ S_inv_rootr%r's rr r sJ c5<< ( (@c LMM c5 ! !6tCykBCC h % %8h8HIJJ c3  B49+NOO nd + +G^H\G]^__xxzH c(msc(m^3BCM>BRRSTWX`TadeTeSffqruqv w   Qw(mc! 5<<3/cAgs8}1UVWI 99ellC5&99%EFMMOI++i(K ALL!C8!GK3H!HIMEJJ}5::<=LKQED <<L) *D$/$7$78I$JT$QO      1CE!a%L E!Hnc"GAq! "aA${{1s73JffZ!##%/0A$(Euzz!c'*QSSU23 dE>rc[U[R5(d[S[ U535e[U[ 5(d[S[ U535e[U[ 5(d[S[ U535e[U[5(d[S[ U535e[XX#S5upEn[XXQ5nU$)aApply ZCA whitening transform. See :class:`~kornia.color.ZCAWhitening` for details. Args: inp: input data torch.Tensor. dim: Specifies the dimension that serves as the samples dimension. unbiased: Whether to use the unbiased estimate of the covariance matrix. eps: a small number used for numerical stability. Returns: Whiten Input data. .. note:: See a working example `here `__. Examples: >>> x = torch.tensor([[0,1],[1,0],[-1,0]], dtype = torch.float32) >>> zca_whiten(x) tensor([[ 0.0000, 1.1547], [ 1.0000, -0.5773], [-1.0000, -0.5773]]) rGrHrIrJF) rQr!rAr2rRr?r@r>r r )rDrrr transformr&rl inp_whitens rr r s4 c5<< ( (@c LMM c5 ! !6tCykBCC h % %8h8HIJJ c3  B49+NOO!#H5AIQ!#$>> # Example where dim = 3 >>> inp = torch.ones((10,3,4,5)) >>> transform_mat = torch.ones((10*3*4,10*3*4)) >>> mean = 2*torch.ones((1,10*3*4)) >>> out = linear_transform(inp, transform_mat, mean, 3) >>> print(out.shape, out.unique()) # Should a be (10,3,4,5) torch.tensor of -120s torch.Size([10, 3, 4, 5]) tensor([-120.]) >>> # Example where dim = 0 >>> inp = torch.ones((10,2)) >>> transform_mat = torch.ones((2,2)) >>> mean = torch.zeros((1,2)) >>> out = linear_transform(inp, transform_mat, mean) >>> print(out.shape, out.unique()) # Should a be (10,2) torch.tensor of 2s torch.Size([10, 2]) tensor([2.]) rKrLrMrNrNrP)rSrTrUr!rVrWrXargsortrYr>r[r\rZr]r3)rDr rrrarbpermperm_invrc inv_orderrfrgrdinp_flat inp_centerinp_transformeds rr r AsXxxzH c(msc(m^3BCM>BRRSTWX`TadeTeSffqruqv w   Qw(mc! 5<<3/cAgs8}1UVWI 99ellC5)95 6D}}T"H;;=I#??,ILL!C8!GK3H!HIMEJJ}5::<=L++i(K""B #56H'J mm$45O%--k.>.>.@AO%--i8O r)rTr8F)rTr8)r)typingrrrr!rkornia.core.utilsr__all__ModulerrAr>r@r?r r r rrr~s$)( - HS299Sngl_ __/3_AF__c_ 5<<x '= =>_D*ELL*s*$*E*]b]i]i*\^_K K).KDILLKWZK \\Kr