
    Ǆg                     d    d dl mZ d dlZd dlmZ d dlmZ d dlmZ d dl	m
Z
 dgZ G d de      Zy)	    )NumberN)constraints)Distribution)broadcast_all)_sizeLaplacec                       e Zd ZdZej
                  ej                  dZej
                  ZdZ	e
d        Ze
d        Ze
d        Ze
d        Zd fd	Zd fd		Z ej$                         fd
edej(                  fdZd Zd Zd Zd Z xZS )r   a  
    Creates a Laplace distribution parameterized by :attr:`loc` and :attr:`scale`.

    Example::

        >>> # xdoctest: +IGNORE_WANT("non-deterministic")
        >>> m = Laplace(torch.tensor([0.0]), torch.tensor([1.0]))
        >>> m.sample()  # Laplace distributed with loc=0, scale=1
        tensor([ 0.1046])

    Args:
        loc (float or Tensor): mean of the distribution
        scale (float or Tensor): scale of the distribution
    )locscaleTc                     | j                   S Nr
   selfs    c/home/mcse/projects/flask_80/flask-venv/lib/python3.12/site-packages/torch/distributions/laplace.pymeanzLaplace.mean!       xx    c                     | j                   S r   r   r   s    r   modezLaplace.mode%   r   r   c                 >    d| j                   j                  d      z  S N   )r   powr   s    r   variancezLaplace.variance)   s    4::>>!$$$r   c                      d| j                   z  S )Ng;f?)r   r   s    r   stddevzLaplace.stddev-   s    $**$$r   c                     t        ||      \  | _        | _        t        |t              r%t        |t              rt        j                         }n| j                  j                         }t        | %  ||       y )Nvalidate_args)
r   r
   r   
isinstancer   torchSizesizesuper__init__)r   r
   r   r    batch_shape	__class__s        r   r&   zLaplace.__init__1   sW    ,S%8$*c6"z%'@**,K((--/KMBr   c                 *   | j                  t        |      }t        j                  |      }| j                  j                  |      |_        | j                  j                  |      |_        t        t        |#  |d       | j                  |_	        |S )NFr   )
_get_checked_instancer   r"   r#   r
   expandr   r%   r&   _validate_args)r   r'   	_instancenewr(   s       r   r+   zLaplace.expand9   st    (()<jj-((//+.JJ%%k2	gs$[$F!00
r   sample_shapereturnc                    | j                  |      }t        j                  | j                  j                        }t        j
                  j                         rt        j                  || j                  j                  | j                  j                        dz  dz
  }| j                  | j                  |j                         z  t        j                  |j                         j                  |j                               z  z
  S | j                  j                  |      j!                  |j"                  dz
  d      }| j                  | j                  |j                         z  t        j                  |j                                z  z
  S )N)dtypedevicer      )min)_extended_shaper"   finfor
   r2   _C_get_tracing_staterandr3   r   signlog1pabsclamptinyr.   uniform_eps)r   r/   shaper7   us        r   rsamplezLaplace.rsampleB   s   $$\2DHHNN+88&&(

5txxORSSVWWA88djj16683ekk5::..7    HHLL((Q: xx$**qvvx/%++quuwh2GGGGr   c                     | j                   r| j                  |       t        j                  d| j                  z         t        j
                  || j                  z
        | j                  z  z
  S r   )r,   _validate_sampler"   logr   r=   r
   r   values     r   log_probzLaplace.log_probP   sS    !!%(		!djj.))EIIedhh6F,G$**,TTTr   c                     | j                   r| j                  |       dd|| j                  z
  j                         z  t	        j
                  || j                  z
  j                          | j                  z        z  z
  S )N      ?)r,   rF   r
   r;   r"   expm1r=   r   rH   s     r   cdfzLaplace.cdfU   sp    !!%(SEDHH,2244u{{dhh##%%

28
 
 
 	
r   c                     |dz
  }| j                   | j                  |j                         z  t        j                  d|j                         z        z  z
  S )NrL   )r
   r   r;   r"   r<   r=   )r   rI   terms      r   icdfzLaplace.icdf\   sA    s{xx$**{{}4u{{2
?7SSSSr   c                 L    dt        j                  d| j                  z        z   S )Nr4   r   )r"   rG   r   r   s    r   entropyzLaplace.entropy`   s    599Q^,,,r   r   )__name__
__module____qualname____doc__r   realpositivearg_constraintssupporthas_rsamplepropertyr   r   r   r   r&   r+   r"   r#   r   TensorrD   rJ   rN   rR   rT   __classcell__)r(   s   @r   r   r      s     *..9M9MNOGK    % % % %C -7EJJL HE HU\\ HU

T-r   )numbersr   r"   torch.distributionsr    torch.distributions.distributionr   torch.distributions.utilsr   torch.typesr   __all__r    r   r   <module>rh      s.      + 9 3  +S-l S-r   