3j, SrSSKJrJrJr SSKrSSKJr SSKJr /SQr "SS\R5r S \RS \RS \RS \R4S jr "SS\R5rS \RS \\R\4S \\R\4S \R4Sjr\SS\RS\S\S\S \R4 Sjj5rg)z=nn.Module containing functionals for intensity normalisation.)ListTupleUnionN)nn)perform_keep_shape_image) Denormalize Normalize denormalize normalizenormalize_min_maxc^\rSrSrSrS\\R\\ \ \ \ 4S\\R\\ \ \ \ 4SS4U4Sjjr S\RS\R4S jr S\ 4S jrS rU=r$) r a"Normalize a torch.Tensor image with mean and standard deviation. .. math:: \text{input[channel] = (input[channel] - mean[channel]) / std[channel]} Where `mean` is :math:`(M_1, ..., M_n)` and `std` :math:`(S_1, ..., S_n)` for `n` channels, Args: mean: Mean for each channel. std: Standard deviations for each channel. Shape: - Input: Image torch.Tensor of size :math:`(*, C, ...)`. - Output: Normalised torch.Tensor with same size as input :math:`(*, C, ...)`. Examples: >>> x = torch.rand(1, 4, 3, 3) >>> out = Normalize(0.0, 255.)(x) >>> out.shape torch.Size([1, 4, 3, 3]) >>> x = torch.rand(1, 4, 3, 3) >>> mean = torch.zeros(4) >>> std = 255. * torch.ones(4) >>> out = Normalize(mean, std)(x) >>> out.shape torch.Size([1, 4, 3, 3]) meanstdreturnNc>[TU]5 [U[[45(a[ R "U/5n[U[[45(a[ R "U/5n[U[[45(a[ R "U5Sn[U[[45(a[ R "U5SnXl X l gN) super__init__ isinstanceintfloattorchtensortuplelistrrselfrr __class__s R/home/wildlama/miniconda3/lib/python3.13/site-packages/kornia/enhance/normalize.pyrNormalize.__init__=s  dS%L ) )<<'D cC< ( (,,u%C dUDM * *<<%d+D cE4= ) ),,s#D)C inputcB[XRUR5$r)r rrrr#s r forwardNormalize.forwardSs 48844r"cnSURSURS3nURRU-$Nz(mean=z, std=)rrr__name__rreprs r __repr__Normalize.__repr__V3 {& !4~~&&--r"rr)r, __module__ __qualname____firstlineno____doc__rrTensorrrrrr&strr/__static_attributes__ __classcell__rs@r r r s<ELL%,U UBC5<<utE{EA B  ,5U\\5ell5.#..r"r datarrrcURn[RR5(a[ U[R 5(a[ U[R 5(d [ S5eURSS:wdURSS:wa&[ SURSURS35eGO[ U[5(a1[R"U/US-URURS9n[ U[5(a1[R"U/US-URURS9nUR(ayURSS:wafURSURS:waFURSS URSS :wa&[ S URS URS 35eUR(ayURSS:wafURSURS:waFURSS URSS :wa&[ S URS URS 35e[R"XRURS9n[R"X RURS9nUSnUSnURUSUSS5U- U- nURU5$)a9Normalize an image/video torch.Tensor with mean and standard deviation. .. math:: \text{input[channel] = (input[channel] - mean[channel]) / std[channel]} Where `mean` is :math:`(M_1, ..., M_n)` and `std` :math:`(S_1, ..., S_n)` for `n` channels, Args: data: Image torch.Tensor of size :math:`(B, C, *)`. mean: Mean for each channel. std: Standard deviations for each channel. Return: Normalised torch.Tensor with same size as input :math:`(B, C, *)`. Examples: >>> x = torch.rand(1, 4, 3, 3) >>> out = normalize(x, torch.tensor([0.0]), torch.tensor([255.])) >>> out.shape torch.Size([1, 4, 3, 3]) >>> x = torch.rand(1, 4, 3, 3) >>> mean = torch.zeros(4) >>> std = 255. * torch.ones(4) >>> out = normalize(x, mean, std) >>> out.shape torch.Size([1, 4, 3, 3]) 6Only torch.Tensor is accepted when converting to ONNX.rzqBatch dimension must be one for broadcasting when converting to ONNX.Try changing mean shape and std shape from (z, z) to (1, C) or (1, C, 1, 1).devicedtypeN5mean length and number of channels do not match. Got  and .4std length and number of channels do not match. Got ).N) shaperonnxis_in_onnx_exportrr7 ValueErrorrrrArB as_tensorview)r<rrrIouts r r r [sj< JJE zz##%%$ --ZU\\5R5RUV V ::a=A 1!2??Czzl"SYYKWsu "3 dE " "<<q 1$++TZZXD c5 ! !,,uuQx/ 4::VC ::$**Q-1,zz!} 1 -$**Ra.DJJrPQN2R #XY]YcYcXddijnjtjtiuuv!wxx 991*yy|tzz!},2A$**Ra.1P #WX[XaXaWbbghlhrhrgsst!uvvtKKtzzJooc++TZZH  ?D i.C58U1Xr:TASHC 88E?r"c^\rSrSrSrS\\R\4S\\R\4SS4U4Sjjr S\RS\R4S jr S\ 4S jr S r U=r$) ra6Denormalize a torch.Tensor image with mean and standard deviation. .. math:: \text{input[channel] = (input[channel] * std[channel]) + mean[channel]} Where `mean` is :math:`(M_1, ..., M_n)` and `std` :math:`(S_1, ..., S_n)` for `n` channels, Args: mean: Mean for each channel. std: Standard deviations for each channel. Shape: - Input: Image torch.Tensor of size :math:`(*, C, ...)`. - Output: Denormalised torch.Tensor with same size as input :math:`(*, C, ...)`. Examples: >>> x = torch.rand(1, 4, 3, 3) >>> out = Denormalize(0.0, 255.)(x) >>> out.shape torch.Size([1, 4, 3, 3]) >>> x = torch.rand(1, 4, 3, 3, 3) >>> mean = torch.zeros(1, 4) >>> std = 255. * torch.ones(1, 4) >>> out = Denormalize(mean, std)(x) >>> out.shape torch.Size([1, 4, 3, 3, 3]) rrrNc:>[TU]5 XlX lgr)rrrrrs r rDenormalize.__init__s  r"r#cB[XRUR5$r)r rrr%s r r&Denormalize.forwards5))TXX66r"cnSURSURS3nURRU-$r)r+r-s r r/Denormalize.__repr__r1r"r2)r,r3r4r5r6rrr7rrr&r8r/r9r:r;s@r rrsl<U5<<#67eELLRWDW>X]a 7U\\7ell7.#..r"rcURn[RR5(a|[ U[R 5(a[ U[R 5(d [ S5eURSS:wdURSS:wa [ S5eGO[ U[5(a1[R"U/US-URURS9n[ U[5(a1[R"U/US-URURS9nUR(ayURSS:wafURSURS:waFURSSURSS:wa&[ S URS URS 35eUR(ayURSS:wafURSURS:waFURSSURSS:wa&[ S URS URS 35e[R"XRURS9n[R"X RURS9nUR5S:Xa*UR"SS /S/UR5S- -Q76nOa[UR5UR5:a:URS 5n[UR5UR5:aM:UR5S:Xa*UR"SS /S/UR5S- -Q76nOa[UR5UR5:a:URS 5n[UR5UR5:aM:[R "XU5$)a-Denormalize an image/video torch.Tensor with mean and standard deviation. .. math:: \text{input[channel] = (input[channel] * std[channel]) + mean[channel]} Where `mean` is :math:`(M_1, ..., M_n)` and `std` :math:`(S_1, ..., S_n)` for `n` channels, Args: data: Image torch.Tensor of size :math:`(B, C, *)`. mean: Mean for each channel. std: Standard deviations for each channel. Return: Denormalised torch.Tensor with same size as input :math:`(B, C, *)`. Examples: >>> x = torch.rand(1, 4, 3, 3) >>> out = denormalize(x, 0.0, 255.) >>> out.shape torch.Size([1, 4, 3, 3]) >>> x = torch.rand(1, 4, 3, 3, 3) >>> mean = torch.zeros(1, 4) >>> std = 255. * torch.ones(1, 4) >>> out = denormalize(x, mean, std) >>> out.shape torch.Size([1, 4, 3, 3, 3]) r>rr?zEBatch dimension must be one for broadcasting when converting to ONNX.r@NrCrDrErFrGrH)rIrrJrKrr7rLrrrArBrMdimrNlen unsqueezeaddcmul)r<rrrIs r r r s< JJE zz##%%$ --ZU\\5R5RUV V ::a=A 1!2de e"3 dE " "<<q 1$++TZZXD c5 ! !,,uuQx/ 4::VC ::$**Q-1,zz!} 2.4::bq>TZZPRQR^3S #XY]YcYcXddijnjtjtiuuv!wxx 991*yy|tzz"~-#))BQ-4::bq>2Q #WX[XaXaWbbghlhrhrgsst!uvvtKKtzzJooc++TZZH xxzQyyB:1#a"8:$**o *>>"%D$**o * wwyA~hhq"8txxzA~ 68#))ntxxz)--#C#))ntxxz) ==S ))r"xmin_valmax_valepsc[U[R5(d[S[ U5S35e[U[ 5(d[S[ U5S35e[U[ 5(d[S[ U5S35eUR nUSUSpeURXVS5nURSSS 9SnURSSS 9Sn X!- Xx- -X- U-- U-n U RU5$) aENormalise an image/video torch.Tensor by MinMax and re-scales the value between a range. The data is normalised using the following formulation: .. math:: y_i = (b - a) * \frac{x_i - \text{min}(x)}{\text{max}(x) - \text{min}(x)} + a where :math:`a` is :math:`\text{min_val}` and :math:`b` is :math:`\text{max_val}`. Args: x: The image torch.Tensor to be normalised with shape :math:`(*, C, H, W)`. min_val: The minimum value for the new range. max_val: The maximum value for the new range. eps: Float number to avoid zero division. Returns: The normalised image torch.Tensor with same shape as input :math:`(*, C, H, W)`. Example: >>> x = torch.rand(1, 5, 3, 3) >>> x_norm = normalize_min_max(x, min_val=-1., max_val=1.) >>> x_norm.min() tensor(-1.) >>> x_norm.max() tensor(1.0000) z$data should be a torch.Tensor. Got: rFz"'min_val' should be a float. Got: z"'max_val' should be a float. Got: rr?rHT)keepdim) rrr7 TypeErrortyperrIrNminmax) r^r_r`rarIBC x_reshapedx_minx_maxx_outs r r r s: a & &>tAwiqIJJ gu % %)r6typingrrrrrkornia.image.utilsr__all__Moduler r7r rrr r r"r rss $D%% 7 W:. :.zAELLA A5<<AELLAH*."))*.ZF*ellF*% e0C*DF*5QVQ]Q]_dQdKeF*jojvjvF*R---e-X]-iniuiu--r"