o "jn%@srddlZddlZddlmZddlmZmZddlmZddl m Z ddl m Z ddl mZGdd d e jZdS) N)_C_ops) check_type convert_dtype)Variable) distribution)in_dynamic_mode)randomcsDeZdZdZdfdd ZdddZdd Zd d Zd d ZZ S)Uniformaj Uniform distribution with `low` and `high` parameters. Mathematical Details The probability density function (pdf) is .. math:: pdf(x; a, b) = \frac{1}{Z}, \ a <=x >> import paddle >>> from paddle.distribution import Uniform >>> paddle.seed(2023) >>> # Without broadcasting, a single uniform distribution [3, 4]: >>> u1 = Uniform(low=3.0, high=4.0) >>> # 2 distributions [1, 3], [2, 4] >>> u2 = Uniform(low=[1.0, 2.0], high=[3.0, 4.0]) >>> # 4 distributions >>> u3 = Uniform(low=[[1.0, 2.0], [3.0, 4.0]], ... high=[[1.5, 2.5], [3.5, 4.5]]) ... >>> # With broadcasting: >>> u4 = Uniform(low=3.0, high=[5.0, 6.0, 7.0]) >>> # Complete example >>> value_tensor = paddle.to_tensor([0.8], dtype="float32") >>> uniform = Uniform([0.], [2.]) >>> sample = uniform.sample([2]) >>> # a random tensor created by uniform distribution with shape: [2, 1] >>> entropy = uniform.entropy() >>> print(entropy) Tensor(shape=[1], dtype=float32, place=Place(cpu), stop_gradient=True, [0.69314718]) >>> lp = uniform.log_prob(value_tensor) >>> print(lp) Tensor(shape=[1], dtype=float32, place=Place(cpu), stop_gradient=True, [-0.69314718]) >>> p = uniform.probs(value_tensor) >>> print(p) Tensor(shape=[1], dtype=float32, place=Place(cpu), stop_gradient=True, [0.50000000]) Nc sntst|dtttjtttfdt|dtttjtttfdd|_ d|_ |dur+|nd|_ d|_ t |tr:t|}t |trCt|}|||rV||_||_t|j |_ nWt |trct |trcd|_ t |tjrut|j dvru|j |_ nt |tjrt|j dvr|j |_ |||\|_|_|j t|jj krtj|j|j d|_tj|j|j d|_t|jjdS) Nlowr highFfloat32T)r Zfloat64dtype)rrintfloatnpZndarrayrlisttupleall_arg_is_floatZbatch_size_unknownnamer isinstanceZ_validate_argsr r rstrZ _to_tensorpaddlecastsuper__init__shape)selfr r r __class__\/var/www/html/Deteccion_Ine/venv/lib/python3.10/site-packages/paddle/distribution/uniform.pyrgsH    zUniform.__init__rc CsJtst|dtdt|dtd|jd}t|j|jj}d|vrr||}t||}t|j|jd |d<t |d|j }t j ||j|j dd|d }t||} t||} | | |j|j} tj| |j|d } | S||}tj||j dd|d tj||j d |j|j} tj| |j|d } |jrtj| ||d S| S) a Generate samples of the specified shape. Args: shape (list): 1D `int32`. Shape of the generated samples. seed (int): Python integer number. Returns: Tensor, A tensor with prepended dimensions shape. The data type is float32. rsampleseed_samplergg?)rminmaxr#rr )rrrrrr r rritemfullrrZuniform_random_batch_size_likeZreshapeadduniformZzerosr) rrr#rZ batch_shapeZ output_shapeZ fill_shapeZzero_tmpZuniform_random_tmpZzero_tmp_reshapeZuniform_random_tmp_reshapeoutputr r r!r"sN      zUniform.samplecCs||j|}tr2|j|k}||jk}t||j}t||j}t||t|j|jS|j d}|j|k}||jk}tj||jd}tj||jd}tj t||t|j|j|dS)zLog probability density/mass function. Args: value (Tensor): The input tensor. Returns: Tensor, log probability.The data type is same with value. Z _log_probr r() _check_values_dtype_in_probsr rr rrrrlogrsubtractrvalueZlb_boolZub_boolZlbZubrr r r!log_probs       zUniform.log_probcCs||j|}tr,|j|k}||jk}t||j}t||j}|||j|jS|jd}|j|k}||jk}tj||jd}tj||jd}tj |||j|j|dS)zProbability density/mass function. Args: value (Tensor): The input tensor. Returns: Tensor, probability. The data type is same with value. Z_probsr r() r.r rr rrrrrdivider1r r r!probss      z Uniform.probscCs |jd}tj|j|j|dS)zShannon entropy in nats. The entropy is .. math:: entropy(low, high) = \\log (high - low) Returns: Tensor, Shannon entropy of uniform distribution.The data type is float32. Z_entropyr()rrr/r r )rrr r r!entropys zUniform.entropy)N)r) __name__ __module__ __qualname____doc__rr"r3r5r6 __classcell__r r rr!r sL 15r )numpyrrrZpaddle.base.data_feederrrZpaddle.base.frameworkrZpaddle.distributionrZpaddle.frameworkrZ paddle.tensorr Distributionr r r r r!s