o p i@sddlmZddlmZddlZddlmZddlmZmZddl m Z ddl m Z ddl mZer@dd lmZdd lmZmZGd d d e jZdS) ) annotations) TYPE_CHECKINGN) distribution) check_type convert_dtype)Variable)exponential_family)in_dynamic_mode)Sequence)TensordtypecseZdZUdZded<ded<ded<d%fd d Zed&d d Zed&d dZd'ddZ d'ddZ d&ddZ gfd(ddZ gfd(ddZ d)ddZd&dd Zd*d#d$ZZS)+Gammaa Gamma distribution parameterized by :attr:`concentration` (aka "alpha") and :attr:`rate` (aka "beta"). The probability density function (pdf) is .. math:: f(x; \alpha, \beta, x > 0) = \frac{\beta^{\alpha}}{\Gamma(\alpha)} x^{\alpha-1}e^{-\beta x} \Gamma(\alpha)=\int_{0}^{\infty} x^{\alpha-1} e^{-x} \mathrm{~d} x, (\alpha>0) Args: concentration (float|Tensor): Concentration parameter. It supports broadcast semantics. The value of concentration must be positive. When the parameter is a tensor, it represents multiple independent distribution with a batch_shape(refer to :ref:`api_paddle_distribution_Distribution`). rate (float|Tensor): Rate parameter. It supports broadcast semantics. The value of rate must be positive. When the parameter is tensor, it represent multiple independent distribution with a batch_shape(refer to :ref:`api_paddle_distribution_Distribution`). Example: .. code-block:: python >>> import paddle >>> # scale input >>> gamma = paddle.distribution.Gamma(0.5, 0.5) >>> print(gamma.mean) Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True, 1.) >>> print(gamma.variance) Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True, 2.) >>> print(gamma.entropy()) Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True, 0.78375685) >>> # tensor input with broadcast >>> gamma = paddle.distribution.Gamma(paddle.to_tensor([0.2, 0.4]), paddle.to_tensor(0.6)) >>> print(gamma.mean) Tensor(shape=[2], dtype=float32, place=Place(gpu:0), stop_gradient=True, [0.33333331, 0.66666663]) >>> print(gamma.variance) Tensor(shape=[2], dtype=float32, place=Place(gpu:0), stop_gradient=True, [0.55555552, 1.11111104]) >>> print(gamma.entropy()) Tensor(shape=[2], dtype=float32, place=Place(gpu:0), stop_gradient=True, [-1.99634242, 0.17067254]) r concentrationrater float | TensorreturnNonecstst|dtttjjfdt|dtttjjfd|||r.||_||_ t |j |_ n| ||\|_|_ t |_ t|jjdS)Nrr r)r rfloatrpaddleZpirValueZ_validate_argsrrrr Z _to_tensorZget_default_dtypesuper__init__shape)selfrr __class__`/home/app/PaddleOCR-VL/.venv_paddleocr/lib/python3.10/site-packages/paddle/distribution/gamma.pyr\s,    zGamma.__init__cCs |j|jS)zVMean of gamma distribution. Returns: Tensor: mean value. rrrrrrmeanzs z Gamma.meancCs|j|jdS)z^Variance of gamma distribution. Returns: Tensor: variance value. )rrpowrrrrvarianceszGamma.variancevaluecCst||S)zProbability density function evaluated at value Args: value (float|Tensor): Value to be evaluated. Returns: Tensor: Probability. )rexplog_probrr$rrrprobs z Gamma.probcCs<|jt|j|jdt||j|t|jS)zLog probability density function evaluated at value Args: value (float|Tensor): Value to be evaluated Returns: Tensor: Log probability. )rrlogrlgammar'rrrr&s  zGamma.log_probcCs4|jt|jt|jd|jt|jS)zUEntropy of gamma distribution Returns: Tensor: Entropy. g?)rrr*rr+digammarrrrentropys  z Gamma.entropyr Sequence[int]cCs6t ||WdS1swYdS)zGenerate samples of the specified shape. Args: shape (Sequence[int], optional): Shape of the generated samples. Returns: Tensor, A tensor with prepended dimensions shape.The data type is float32. N)rZno_gradrsamplerrrrrsamples $z Gamma.samplecCs.tjj||d}t|j||j|S)aGenerate reparameterized samples of the specified shape. Args: shape (Sequence[int], optional): Shape of the generated samples. Returns: Tensor: A tensor with prepended dimensions shape.The data type is float32. )Z sample_shape)r DistributionZ _extend_shaperZstandard_gammarexpandrr0rrrr/s   z Gamma.rsampleothercCst|tstdt||jt|j|j}t|jt|j}|j|jt |j}|j|j|j|j}||||S)zThe KL-divergence between two gamma distributions. Args: other (Gamma): instance of Gamma. Returns: Tensor: kl-divergence between two gamma distributions. z/Expected type of other is Exponential, but got ) isinstancer TypeErrortyperrr*rr+r,)rr4t1t2t3Zt4rrr kl_divergences  zGamma.kl_divergencecCs|jd|j fSNr)rrrrr_natural_parametersszGamma._natural_parametersxycCs&t|d|dt| Sr<)rr+r*Z reciprocal)rr>r?rrr_log_normalizers&zGamma._log_normalizer)rrrrrr)rr )r$rrr )rr.rr )r4r rr )r>r r?r rr )__name__ __module__ __qualname____doc____annotations__rpropertyr r#r(r&r-r1r/r;r=r@ __classcell__rrrrr s$ 7       r ) __future__rtypingrrrZpaddle.base.data_feederrrZpaddle.base.frameworkrZpaddle.distributionrZpaddle.frameworkr collections.abcr r r ZExponentialFamilyr rrrrs