[ThSSKJr SSKrSSKJrJrJr SSKrSSKJ r SSK J r J r JrJrJr \(aSSKJr S/r\ R(r\R*"S 5S S j5rS S jrg)) annotationsN)Callable TYPE_CHECKINGUnion)dim) _ellipsis AnonymousAxiscomma_separate parse_patternvalidate_rearrange_expressions)Sequence rearrangec ^^^[X5up4[X4U5 [SUR55nUR(aOU[ UR5S- - n[ UR 5S- nXW-=o:a[SUSUS35eOCSn[ UR 5n[ UR5=o:wa[SUSUS35eXv-U-n U S:XaS$[S [U 555m0m/n SmURHn [U [5(atU H(n [U [5(deTT4TU 'TS- mM* U (d=[S 5n TT4TU 'U RU 5 U RU 5 TS- mMMU [:Xa/[n [UU4S j[U555TU 'TU- mM[S U 35e SU4S jjnU"UR5nU"UR5n[U4SjU 55n[U4SjUR!555nSnSUS[#T5SU S3U(aSR%SU55OS-S[#U5S[#U5S3-U(aS[#U/5S3OS-n['U5 [)5U$)aVTranslate an `einops`-style pattern into a callable that performs the rearrange using first-class dimensions. Since the an equivalent result is computed for tensors with the same number of dimensions, with the same pattern and specified axes lengths, this function can be memoized. Args: tensor_ndim (int): the number of dimensions in the tensor to rearrange pattern (str): the `einops`-style rearrangement pattern axes_lengths (int): any additional length specifications for dimensions Returns: Callable[[torch.Tensor], torch.Tensor]: a callable that performs the rearrangement c3.# UH o(+v M g7fN).0rs R/var/www/auris/envauris/lib/python3.13/site-packages/functorch/einops/rearrange.py -_create_rearrange_callable...s:)9#gg)9srz!Number of dimensions in pattern (zH) must be less than or equal to the number of dimensions in the tensor ()rz;) must be equal to the number of dimensions in the tensor (cU$rr)tensors r,_create_rearrange_callable..Esfc3,# UH nSU3v M g7f)dNr)ris rrrGs-M}!!g}s1c34># UH nTTU-v M g7frr)rjdims_ifirst_class_dimss rrr]s36L !,6LUnexpected dimension: c>/nUHnn[U[5(a&UR[U4SjU555 M>U[:XaUR T[5 Mb[ SU35e U$)ztConvert a `ParsedExpression.composition` into a `Tensor.__getitem__` index of strings representing first class dims.c3B># UHnTUHnUv M M g7frr)r identifierridentifier_dim_maps rrJ_create_rearrange_callable..composition_to_dims..ms**3J#5j#AC#A*3sr() isinstancelistappendtupler extend ValueError) compositiondim_composition dimensionr,s rcomposition_to_dims7_create_rearrange_callable..composition_to_dimsds~ >@$I)T**&&*3i'&&'9)'DE #9)!EFF%rc34># UH nTUSv M g7frNr)raxisr,s rrr{sHid(.q1ir'c3<># UHupTUSU4v M g7fr:r)rr;lengthr,s rrr|s'BV,$ D !! $f-BVs do_rearrangezdef z(tensor): z = dims(z) c38# UHupSUSUS3v M g7f)z z.size =  Nr)rrr=s rrrs%DU=C$se8F82.DUsz tensor = tensor[z].order(z return tensor.sum(z, keepdim=False) z return tensor )r4z5Sequence[Union[list[Union[str, AnonymousAxis]], str]]returnz!list[Union[str, tuple[str, ...]]])r r sumr4 has_ellipsislen identifiersr3r1ranger.r/strr r0r itemsr joinexeclocals) tensor_ndimpattern axes_lengthsleftright n_anon_dimsn_ellipsis_dims n_named_dims pattern_ndimn_dims anon_axesr6r+ anon_axisr7 left_dims right_dims anon_dimsspecified_lengthscustom_rearrange_callable_namecustom_rearrange_callable_coder%r&r,s @@@r_create_rearrange_callabler_sY" 6KD"4 =:)9)9::K %T-=-=)>)BC4++,q0 '6 6L E3L>B--8M<  F 4++,  0 01 1L A3L>B*m1.  +k 9F {$$(--MuV}-M(MKM%'IF%% i & &' !*c22222B62J1L":.!  ( )#. 1A&1I0K"9-  +  +!  ) #"J-236;O6L3. z * o %F5i[AB B+&.J **$D$4$45I$U%6%67JHiHHIBNBTBTBV&4"123!"234HVHC I! GGDU   ! !: ;8NS]D^C__b c d%^YK%@$AAS T& #( '( 82 33rc [U[R5(d[R"U5n[ UR U40UD6nU"U5$)a2A native implementation of `einops.rearrange`, a reader-friendly smart element reordering for multidimensional tensors. This operation includes functionality of transpose (axes permutation), reshape (view), squeeze, unsqueeze, stack, concatenate and other operations. See: https://einops.rocks/api/rearrange/ Args: tensor (Tensor or sequence of Tensor): the tensor(s) to rearrange pattern (str): the rearrangement pattern axes_lengths (int): any additional length specifications for dimensions Returns: Tensor: the rearranged tensor Examples: >>> # suppose we have a set of 32 images in "h w c" format (height-width-channel) >>> images = torch.randn((32, 30, 40, 3)) >>> # stack along first (batch) axis, output is a single array >>> rearrange(images, "b h w c -> b h w c").shape torch.Size([32, 30, 40, 3]) >>> # concatenate images along height (vertical axis), 960 = 32 * 30 >>> rearrange(images, "b h w c -> (b h) w c").shape torch.Size([960, 40, 3]) >>> # concatenated images along horizontal axis, 1280 = 32 * 40 >>> rearrange(images, "b h w c -> h (b w) c").shape torch.Size([30, 1280, 3]) >>> # reordered axes to "b c h w" format for deep learning >>> rearrange(images, "b h w c -> b c h w").shape torch.Size([32, 3, 30, 40]) >>> # flattened each image into a vector, 3600 = 30 * 40 * 3 >>> rearrange(images, "b h w c -> b (c h w)").shape torch.Size([32, 3600]) >>> # split each image into 4 smaller (top-left, top-right, bottom-left, bottom-right), 128 = 32 * 2 * 2 >>> rearrange(images, "b (h1 h) (w1 w) c -> (b h1 w1) h w c", h1=2, w1=2).shape torch.Size([128, 15, 20, 3]) >>> # space-to-depth operation >>> rearrange(images, "b (h h1) (w w1) c -> b h w (c h1 w1)", h1=2, w1=2).shape torch.Size([32, 15, 20, 12]) )r.torchTensorstackr_ndim)rrNrOrearrange_callables rrrsMf fell + +V$3 W , f %%r)rMintrNrHrOrfrBz&Callable[[torch.Tensor], torch.Tensor])rzAUnion[torch.Tensor, list[torch.Tensor], tuple[torch.Tensor, ...]]rNrHrOrfrBz torch.Tensor) __future__r functoolstypingrrrra functorch._Cr_C_parsingr r r r r collections.abcr__all__dims lru_cacher_rrrrrqs"11 "( - ww S{4{4"{447{4+{4{4|:& M:& :&:& :&r