tj=#ddlZddlmZddlmZddlmZddlmZddlm Z m Z ddl m Z ddl mZd gZGd d eZdS) N)Tensor) constraints) Distribution) Independent)ComposeTransform Transform)_sum_rightmost)_sizeTransformedDistributionc @eZdZUdZiZeeejfe d< dde de e e zde dzddffd Zdfd Zejd d Zede fd ZejfdZejfdedefdZdZdZdZdZxZS)r a Extension of the Distribution class, which applies a sequence of Transforms to a base distribution. Let f be the composition of transforms applied:: X ~ BaseDistribution Y = f(X) ~ TransformedDistribution(BaseDistribution, f) log p(Y) = log p(X) + log |det (dX/dY)| Note that the ``.event_shape`` of a :class:`TransformedDistribution` is the maximum shape of its base distribution and its transforms, since transforms can introduce correlations among events. An example for the usage of :class:`TransformedDistribution` would be:: # Building a Logistic Distribution # X ~ Uniform(0, 1) # f = a + b * logit(X) # Y ~ f(X) ~ Logistic(a, b) base_distribution = Uniform(0, 1) transforms = [SigmoidTransform().inv, AffineTransform(loc=a, scale=b)] logistic = TransformedDistribution(base_distribution, transforms) For more examples, please look at the implementations of :class:`~torch.distributions.gumbel.Gumbel`, :class:`~torch.distributions.half_cauchy.HalfCauchy`, :class:`~torch.distributions.half_normal.HalfNormal`, :class:`~torch.distributions.log_normal.LogNormal`, :class:`~torch.distributions.pareto.Pareto`, :class:`~torch.distributions.weibull.Weibull`, :class:`~torch.distributions.relaxed_bernoulli.RelaxedBernoulli` and :class:`~torch.distributions.relaxed_categorical.RelaxedOneHotCategorical` arg_constraintsNbase_distribution transforms validate_argsreturncZt|tr |g|_nWt|tr0t d|Dst d||_nt d||j|jz}t|j}t|j}t||j j kr t d|j j d|d| |}| |}||kr/|dt||z } || }|j j |z } | dkrt|| }||_|jj |j j z } t%|jj || z} t|| kr"t'dt|d | t|| z } |d| }|| d}t)||| dS) Nc3@K|]}t|tVdSN) isinstancer).0ts j/var/www/html/CA-Chatbot/venv/lib/python3.11/site-packages/torch/distributions/transformed_distribution.py z3TransformedDistribution.__init__..?s,DDAz!Y//DDDDDDz6transforms must be a Transform or a list of Transformsz0transforms must be a Transform or list, but was z9base_distribution needs to have shape with size at least z , but got .rzforward_shape length z must be >= event_dim r)rrrlistall ValueError batch_shape event_shapelenrdomain event_dim forward_shape inverse_shapeexpandr base_distcodomainmaxAssertionErrorsuper__init__)selfrrr base_shapebase_event_dim transformr%expanded_base_shapebase_batch_shapereinterpreted_batch_ndimstransform_change_in_event_dimr$cutr r! __class__s rr-z TransformedDistribution.__init__4s j) , , DOO D ) ) DDDDDDD  L)DOOO:OO  '25F5RR .:;;$T_55 z??Y-7 7 7IL\Lfr| "// ;; '55mDD , , ,2;#)**^;;  !2 8 89I J J $-$4$>$O! $q ( ( +!#<!! +   (9+;+E E &   ( : :   }   ) ) ]M(:(:]]R[]] -  9,#DSD) #CDD)  kOOOOOrc|t|}tj|}||jz}t |jD]}||}|dt|t|j jz }|j ||_ |j|_tt| ||jd|j |_ |S)NFr)_get_checked_instancer torchSizer!reversedrr&r"r(r'r,r-_validate_args)r.r _instancenewshaperr3r7s rr'zTransformedDistribution.expandps(()@)LLj-- d..$/** + +AOOE**EE !O3u::DN4N0O0O#O!OP--.>??  %s++44 ) 5   "0 rF) is_discretec|js |jjS|jdj}t |j|jkr/tj|t |j|jz }|S)N) rr(supportr)r"r!r$r independent)r.rDs rrDzTransformedDistribution.supportsr *>) )/"%. t 7#4 4 4!-T-..1BBGrc|jjSr)r( has_rsample)r.s rrGz#TransformedDistribution.has_rsamples ~))rctj5|j|}|jD] }||}|cdddS#1swxYwYdS)a Generates a sample_shape shaped sample or sample_shape shaped batch of samples if the distribution parameters are batched. Samples first from base distribution and applies `transform()` for every transform in the list. N)r:no_gradr(samplerr. sample_shapexr1s rrJzTransformedDistribution.samples]__  %%l33A!_ ! ! IaLL                   s1AAArLcd|j|}|jD] }||}|S)a$ Generates a sample_shape shaped reparameterized sample or sample_shape shaped batch of reparameterized samples if the distribution parameters are batched. Samples first from base distribution and applies `transform()` for every transform in the list. )r(rsamplerrKs rrOzTransformedDistribution.rsamples> N " "< 0 0  I ! AArc|jr||t|j}d}|}t |jD]i}||}||jj|j jz z }|t| ||||jjz z }|}j|t|j ||t|j jz z}|S)z Scores the sample by inverting the transform(s) and computing the score using the score of the base distribution and the log abs det jacobian. g)r=_validate_sampler"r!r<rinvr#r$r)r log_abs_det_jacobianr(log_prob)r.valuer$rTyr1rMs rrTz TransformedDistribution.log_probs   )  ! !% ( ( (()) #& !$/22  I a  A )3i6H6RR RI...q!44I,66##HAAn N # #A & & C8R4S4S(S   rc~d}|jD] }||jz} t|tr|dkr|S||dz zdzS)zu This conditionally flips ``value -> 1-value`` to ensure :meth:`cdf` is monotone increasing. g?)rsignrint)r.rUrYr1s r_monotonize_cdfz'TransformedDistribution._monotonize_cdfs[  ) )I).(DD dC  TQYYLus{#c))rc|jdddD]}||}|jr|j||j|}||}|S)z Computes the cumulative distribution function by inverting the transform(s) and computing the score of the base distribution. NrC)rrRr=r(rQcdfr[r.rUr1s rr]zTransformedDistribution.cdfs 2. ) )IMM%((EE   3 N + +E 2 2 2""5))$$U++ rc||}|j|}|jD] }||}|S)z Computes the inverse cumulative distribution function using transform(s) and computing the score of the base distribution. )r[r(icdfrr^s rr`zTransformedDistribution.icdfsS $$U++##E** % %IIe$$EE rr)__name__ __module__ __qualname____doc__r dictstrr Constraint__annotations__rrrboolr-r'dependent_propertyrDpropertyrGr:r;rJr rrOrTr[r]r` __classcell__)r7s@rr r sB:rus ++++++999999777777FFFFFFFF444444 % %RRRRRlRRRRRr