2Vh1ddlmZddlmZddlmZddlmZd dZedGdd eZ ed d d Z y))ops) keras_export)Merge) normalizeNc|j}|j}t|}t|}|dks|dkrtd|d|d|d}|d}||||k7rtd|d|dt|tr||g}||dk(r |dz |dz g}n |dz |dz g}t d |Drtd |d t |}|ddkr |dxx|z cc<|ddkr |dxx|z cc<d|vr td |\} } || } || } | .| ,| | k7r'td |d|d|d|dd|dd| d| d|} |}|dk(r tj|d}| dz } |dz }|dk(rtj|d}|dz }| |dz k7rNt t|}t| |dz D] }||dz||<| |d<tj||}| dk7rLt t|}t| ddD] }||dz ||<| |d<tj||}|dkDr>tj|}|dd}|dd|df}tj||}d}nd}|dkDr>tj|}|dd}|d|ddf}tj||}d}nd}tj||}tj|}d}|r|ddz|ddz}d}|r |ddz}d}|rtj||}| dk(rtj|d}|S|dk(rtj|d}|S)a7Batchwise dot product. `batch_dot` is used to compute dot product of `x` and `y` when `x` and `y` are data in batch, i.e. in a shape of `(batch_size, :)`. `batch_dot` results in a tensor or variable with less dimensions than the input. If the number of dimensions is reduced to 1, we use `expand_dims` to make sure that ndim is at least 2. Shape inference: Let `x`'s shape be `(100, 20)` and `y`'s shape be `(100, 30, 20)`. If `axes` is (1, 2), to find the output shape of resultant tensor, loop through each dimension in `x`'s shape and `y`'s shape: * `x.shape[0]` : 100 : append to output shape * `x.shape[1]` : 20 : do not append to output shape, dimension 1 of `x` has been summed over. (`dot_axes[0]` = 1) * `y.shape[0]` : 100 : do not append to output shape, always ignore first dimension of `y` * `y.shape[1]` : 30 : append to output shape * `y.shape[2]` : 20 : do not append to output shape, dimension 2 of `y` has been summed over. (`dot_axes[1]` = 2) `output_shape` = `(100, 30)` Example: >>> x_batch = np.ones(shape=(32, 20, 1)) >>> y_batch = np.ones(shape=(32, 30, 20)) >>> xy_batch_dot = batch_dot(x_batch, y_batch, axes=(1, 2)) Args: x: Keras tensor or variable with `ndim >= 2`. y: Keras tensor or variable with `ndim >= 2`. axes: Tuple or list of integers with target dimensions, or single integer. The sizes of `x.shape[axes[0]]` and `y.shape[axes[1]]` should be equal. Returns: A tensor with shape equal to the concatenation of `x`'s shape (less the dimension that was summed over) and `y`'s shape (less the batch dimension and the dimension that was summed over). If the final rank is 1, we reshape it to `(batch_size, 1)`. zICannot do batch_dot on inputs with rank < 2. Received inputs with shapes z and .rNzVCannot do batch_dot on inputs with different batch sizes. Received inputs with shapes c3HK|]}t|ttfywN) isinstancelisttuple).0as L/home/dcms/DCMS/lib/python3.12/site-packages/keras/src/layers/merging/dot.py zbatch_dot..Vs 6A:a$ ' 6s "zYMultiple target dimensions are not supported. Expected: None, int, (int, int), Provided:  zCannot perform batch_dot over axis 0. If your inputs are not batched, add a dummy batch dimension to your inputs using keras.ops.expand_dims(x, 0)z*Cannot do batch_dot on inputs with shapes z with axes=z . x.shape[z ] != y.shape[z] (z != z).TF)shapelen ValueErrorr intanyrr expand_dimsrange transposereshapematmulsqueeze)xyaxesx_shapey_shapex_ndimy_ndim x_batch_size y_batch_sizea0a1d1d2 orig_x_ndim orig_y_ndimpatterni x_mid_dimsx_squashed_shape x_squashed y_trail_dimsy_squashed_shape y_squashedresult output_shape do_reshapes r batch_dotr<sSZggGggG \F \F zVaZ+iuWIQ (  1:L1:LL$< < '/)5 ,  $d| | Q;QJ +DQJ +D 6 66a !   :D Aw{ Q6 Aw{ Q6 Dy 7  FB B B ~".R2X8iuWI[7Awi}T!WISDB H  KK{ OOAq ! a!  { OOAq !!  VaZuV}%r6A:& (A QGAJ ( MM!W % QwuV}%r1b! (A QGAJ ( MM!W %z))A,Qr] #AJGBK8 KK+ ,   z))A,qr{ #AJ B7 KK+ ,  ZZ1 F99V$LJ#BQ'*4|BC7HH  #CR(<7  V\2aVQ' M  VR( Mzkeras.layers.DotcHeZdZdZdfd ZdZdZdZd dZfdZ xZ S) DotaComputes element-wise dot product of two tensors. It takes a list of inputs of size 2, and the axes corresponding to each input along with the dot product is to be performed. Let's say `x` and `y` are the two input tensors with shapes `(2, 3, 5)` and `(2, 10, 3)`. The batch dimension should be of same size for both the inputs, and `axes` should correspond to the dimensions that have the same size in the corresponding inputs. e.g. with `axes=(1, 2)`, the dot product of `x`, and `y` will result in a tensor with shape `(2, 5, 10)` Example: >>> x = np.arange(10).reshape(1, 5, 2) >>> y = np.arange(10, 20).reshape(1, 2, 5) >>> keras.layers.Dot(axes=(1, 2))([x, y]) Usage in a Keras model: >>> x1 = keras.layers.Dense(8)(np.arange(10).reshape(5, 2)) >>> x2 = keras.layers.Dense(8)(np.arange(10, 20).reshape(5, 2)) >>> y = keras.layers.Dot(axes=1)([x1, x2]) Args: axes: Integer or tuple of integers, axis or axes along which to take the dot product. If a tuple, should be two integers corresponding to the desired axis from the first input and the desired axis from the second input, respectively. Note that the size of the two selected axes must match. normalize: Whether to L2-normalize samples along the dot product axis before taking the dot product. If set to `True`, then the output of the dot product is the cosine proximity between the two samples. **kwargs: Standard layer keyword arguments. Returns: A tensor, the dot product of the samples from the inputs. c dt|d i|t|tstt|tt fst d|t|dk7rtd|t|dtrt|dtstd|||_ ||_ d|_ d|_ y) NzPInvalid type for argument `axes`: it should be a list or an int. Received: axes=rzSInvalid format for argument `axes`: it should contain two elements. Received: axes=rr zUInvalid format for argument `axes`: list elements should be integers. Received: axes=TF) super__init__r rrr TypeErrorrrr$rsupports_masking_reshape_required)selfr$rkwargs __class__s rrCz Dot.__init__s "6"$$dT5M288E4=9;1$))4 7 QQ >V^  dii %yy1} CK/S[1HI {Q99D $q'?fT!Wo -%d1g/z$q'%$q'?#:d1gY7&&,XRx9  .r=ct|dk7rtd||d}|d}t|jtre|jdkrE|jt|j z|jt|j zg}n|jgdz}ng}t t|jD]j}|j|dkr8|j|j|t||j zM|j|j|l|jr t||d}t||d}t|||}|S)NrzEA `Dot` layer should be called on exactly 2 inputs. Received: inputs=rr )axis) rrr r$rrrappendrr<)rGinputsx1x2r$r2outputs r_merge_functionzDot._merge_function$s> v;! $$*8- AY AY dii %yy1}IIBHH -IIBHH -  {QD3tyy>* .99Q>2DG,B2DG,B2r4( r=cFt|ttfrt|dk7rt d|t|d}t|d}t|j t rQ|j dkr1|j t|z|j t|zg}n|j gdz}n |j }|j|d|j|d|jd||z}t|dk(r|dgz }t|S)NrrKrr )r rrrrr$rpop)rGrLrMrNr$r:s rcompute_output_shapezDot.compute_output_shapeCs+t}5[9IQ9N))4 7 k!n%k!n% dii %yy1} CK/S[1HI {Q99D 47 47 1  |  ! QC L\""r=cyr rA)rGrSmasks r compute_maskzDot.compute_maskZsr=c|j|jd}t| }t t |j t |j zS)N)r$r)r$rrB get_configdictritems)rGconfig base_configrIs rr_zDot.get_config]sPII g(* D**,-V\\^0DDEEr=)Fr ) __name__ __module__ __qualname____doc__rCrOrWrZr]r_ __classcell__)rIs@rr?r?s.'R'.:>#.FFr=r?zkeras.layers.dotc (tdd|i||S)aSFunctional interface to the `Dot` layer. Args: inputs: A list of input tensors (at least 2). axes: Integer or tuple of integers, axis or axes along which to take the dot product. normalize: Whether to L2-normalize samples along the dot product axis before taking the dot product. If set to `True`, then the output of the dot product is the cosine proximity between the two samples. **kwargs: Standard layer keyword arguments. Returns: A tensor, the dot product of the samples from the inputs. r$rA)r?)rSr$rHs rdotrjfs" $3 #D #F #F ++r=r )r) keras.srcrkeras.src.api_exportr#keras.src.layers.merging.base_mergerkeras.src.utils.numerical_utilsrr<r?rjrAr=rrosZ-55{| !]F%]F"]F@ !,",r=