ΑiSSKJr SSKrSSKJr SSKJrJrJr \(aSSKJ r SSK J r SSK J r SSKJr "S S \R5rS S jrg) ) annotationsN) TYPE_CHECKING) distribution independent transform)Sequence)Tensor) Distribution) Transformc|^\rSrSr%SrS\S'S\S'S U4Sjjr/4S Sjjr/4S S jjrSS jr S r U=r $)TransformedDistributiona7 Applies a sequence of Transforms to a base distribution. Args: base (Distribution): The base distribution. transforms (Sequence[Transform]): A sequence of ``Transform`` . Examples: .. code-block:: python >>> import paddle >>> paddle.seed(2023) >>> from paddle.distribution import transformed_distribution >>> d = transformed_distribution.TransformedDistribution( ... paddle.distribution.Normal(0., 1.), ... [paddle.distribution.AffineTransform(paddle.to_tensor(1.), paddle.to_tensor(2.))] ... ) >>> # doctest: +SKIP('random sample') >>> print(d.sample([10])) Tensor(shape=[10], dtype=float32, place=Place(cpu), stop_gradient=True, [ 3.22699189, 1.12264419, 0.50283587, 1.83812487, -2.00740123, -2.70338631, 1.26663208, 4.47909021, -0.11529565, 4.32719326]) >>> print(d.log_prob(paddle.to_tensor(0.5))) Tensor(shape=[], dtype=float32, place=Place(cpu), stop_gradient=True, -1.64333570) >>> # doctest: -SKIP r baseSequence[Transform] transformsc>[U[R5(d[S[ U5S35e[U[ R 5(d[S[ U5S35e[SU55(d [S5e[R"U5nURUR-nXl X l U(d%[TU]=URUR5 g[!URUR-5UR"R$:a/['SUR"R$S[!U5S35eUR"R$[!UR5:a@[(R*"XR"R$[!UR5- 5nUR-URUR-5nUR.R$[1[!UR5UR"R$- S5-n[TU]=US[!U5U- U[!U5U- S5 g) Nz1Expected type of 'base' is Distribution, but got .zGExpected type of 'transforms' is Sequence[Transform] or Chain, but got c3V# UHn[U[R5v M! g7f)N) isinstancerr ).0ts l/var/www/html/banglarbhumi/venv/lib/python3.13/site-packages/paddle/distribution/transformed_distribution.py 3TransformedDistribution.__init__..Ks Jz!:a!4!455zs')z1All element of transforms must be Transform type.z.'base' needs to have shape with size at least z , bug got r)rrr TypeErrortypetypingrallrChainTransform batch_shape event_shape_base _transformssuper__init__len_domain event_rank ValueErrorr Independent forward_shape _codomainmax)selfrrchain base_shapetransformed_shapetransformed_event_rank __class__s rr% TransformedDistribution.__init__@s+$ 9 9::CDJTransformedDistribution.rsample{s: JJ  u %!!A ! A"r5cSnUn[UR5n[UR5HynUR U5nXER R URR - - nU[URU5XER R - 5- nUnM{ U[URRU5U[URR5- 5- nU$)zThe log probability evaluated at value. Args: value (Tensor): The value to be evaluated. Returns: Tensor: The log probability. g) r&r!reversedr#inverser'r(r,_sum_rightmostforward_log_det_jacobianr"log_prob)r.valuerEyr(rr;s rrE TransformedDistribution.log_probs ))* $**+A ! A ))..1G1GG GJ.**1-zIIrE__static_attributes__ __classcell__)r3s@rr r sR> ##+  + .A+ + Z-/ .0 r5r c\US:a%UR[[U*S555$U$)Nr)sumlistrange)rFns rrCrCs),-E599T%A,' (r`s?# DD(=7Bl77BJ=r5