3
O4W^·„  ã               @   sè   d dl Z d dljjZd dlmZ d dlmZ dddddd	d
dddddddddgZ	dd„ Z
d)dd„Zd*dd„Zdd„ Zdd	„ Zdd
„ Zdd„ Zd+dd„Ze `d,dd„Zd-d d„Zd.d"d„Zd#d„ Zd/d%d„Zd&d„ Zd0d'd„Zd(d„ ZdS )1é    N)Úinf)ÚproductÚalign_tensorsÚbroadcast_tensorsÚcartesian_prodÚchain_matmulÚeinsumÚisfiniteÚisinfÚluÚ	lu_unpackÚnormÚmeshgridÚsplitÚstftÚ	tensordotÚuniqueÚunique_consecutivec              G   s   t jjj| ƒS )a   broadcast_tensors(*tensors) -> List of Tensors

    Broadcasts the given tensors according to :ref:`broadcasting-semantics`.

    Args:
        *tensors: any number of tensors of the same type

    .. warning::

        More than one element of a broadcasted tensor may refer to a single
        memory location. As a result, in-place operations (especially ones that
        are vectorized) may result in incorrect behavior. If you need to write
        to the tensors, please clone them first.

    Example::

        >>> x = torch.arange(3).view(1, 3)
        >>> y = torch.arange(2).view(2, 1)
        >>> a, b = torch.broadcast_tensors(x, y)
        >>> a.size()
        torch.Size([2, 3])
        >>> a
        tensor([[0, 1, 2],
                [0, 1, 2]])
    )ÚtorchÚ_CÚ_VariableFunctionsr   )Útensors© r   ú1/tmp/pip-build-e66j97b2/torch/torch/functional.pyr      s    c             C   s   | j ||ƒS )að  Splits the tensor into chunks.

    If :attr:`split_size_or_sections` is an integer type, then :attr:`tensor` will
    be split into equally sized chunks (if possible). Last chunk will be smaller if
    the tensor size along the given dimension :attr:`dim` is not divisible by
    :attr:`split_size`.

    If :attr:`split_size_or_sections` is a list, then :attr:`tensor` will be split
    into ``len(split_size_or_sections)`` chunks with sizes in :attr:`dim` according
    to :attr:`split_size_or_sections`.

    Arguments:
        tensor (Tensor): tensor to split.
        split_size_or_sections (int) or (list(int)): size of a single chunk or
            list of sizes for each chunk
        dim (int): dimension along which to split the tensor.
    )r   )ÚtensorZsplit_size_or_sectionsÚdimr   r   r   r   7   s    Tc             C   sx  | j d	ƒ}|r4| jƒ }| jƒ }|jd
dd�jdƒ nd }}|�rj|d }| jƒ dk� rþtj|| j| j	d�j
| ƒjƒ }xðttdd„ | jdd… ƒŽ D ]d}	tt|ƒƒ}
x2t||	 ƒD ]"\}}|
| |
|  |
|< |
|< q²W ||	 jdtj|
|jd�ƒ||	< q”W njtj|| j| j	d�}tt|ƒƒ}
x0t|ƒD ]$\}}|
| |
|  |
|< |
|< �q(W |jdtj|
|jd�ƒ}nd}|||fS )a·  Unpacks the data and pivots from a LU factorization of a tensor.

    Returns a tuple of tensors as ``(the pivots, the L tensor, the U tensor)``.

    Arguments:
        LU_data (Tensor): the packed LU factorization data
        LU_pivots (Tensor): the packed LU factorization pivots
        unpack_data (bool): flag indicating if the data should be unpacked
        unpack_pivots (bool): flag indicating if the pivots should be unpacked

    Example::

        >>> A = torch.randn(2, 3, 3)
        >>> A_LU, pivots = A.lu()
        >>> P, A_L, A_U = torch.lu_unpack(A_LU, pivots)
        >>>
        >>> # can recover A from factorization
        >>> A_ = torch.bmm(P, torch.bmm(A_L, A_U))
    é   é   )Zdim1Zdim2N)ÚdeviceÚdtypec             S   s   t t| ƒƒS )N)ÚlistÚrange)Úxr   r   r   Ú<lambda>r   s    zlu_unpack.<locals>.<lambda>)r   éÿÿÿÿéþÿÿÿr$   r%   )ÚsizeZtriuZtrilZdiagonalZfill_r   r   Zeyer   r   Z	expand_asÚcloner   ÚmapÚshaper    r!   Ú	enumerateZindex_selectZ	as_tensor)ZLU_dataZ	LU_pivotsZunpack_dataZunpack_pivotsÚszÚUÚLZLU_pivots_zero_idxÚPÚidxZfinal_orderÚkÚjr   r   r   r   P   s,    
"& c             G   s6   t |ƒdkr&t|d ttfƒr&|d }tjjj| |ƒS )aå
  einsum(equation, *operands) -> Tensor

This function provides a way of computing multilinear expressions (i.e. sums of products) using the
Einstein summation convention.

Args:
    equation (string): The equation is given in terms of lower case letters (indices) to be associated
           with each dimension of the operands and result. The left hand side lists the operands
           dimensions, separated by commas. There should be one index letter per tensor dimension.
           The right hand side follows after `->` and gives the indices for the output.
           If the `->` and right hand side are omitted, it implicitly defined as the alphabetically
           sorted list of all indices appearing exactly once in the left hand side.
           The indices not apprearing in the output are summed over after multiplying the operands
           entries.
           If an index appears several times for the same operand, a diagonal is taken.
           Ellipses `...` represent a fixed number of dimensions. If the right hand side is inferred,
           the ellipsis dimensions are at the beginning of the output.
    operands (list of Tensors): The operands to compute the Einstein sum of.

Examples::

    >>> x = torch.randn(5)
    >>> y = torch.randn(4)
    >>> torch.einsum('i,j->ij', x, y)  # outer product
    tensor([[-0.0570, -0.0286, -0.0231,  0.0197],
            [ 1.2616,  0.6335,  0.5113, -0.4351],
            [ 1.4452,  0.7257,  0.5857, -0.4984],
            [-0.4647, -0.2333, -0.1883,  0.1603],
            [-1.1130, -0.5588, -0.4510,  0.3838]])


    >>> A = torch.randn(3,5,4)
    >>> l = torch.randn(2,5)
    >>> r = torch.randn(2,4)
    >>> torch.einsum('bn,anm,bm->ba', l, A, r) # compare torch.nn.functional.bilinear
    tensor([[-0.3430, -5.2405,  0.4494],
            [ 0.3311,  5.5201, -3.0356]])


    >>> As = torch.randn(3,2,5)
    >>> Bs = torch.randn(3,5,4)
    >>> torch.einsum('bij,bjk->bik', As, Bs) # batch matrix multiplication
    tensor([[[-1.0564, -1.5904,  3.2023,  3.1271],
             [-1.6706, -0.8097, -0.8025, -2.1183]],

            [[ 4.2239,  0.3107, -0.5756, -0.2354],
             [-1.4558, -0.3460,  1.5087, -0.8530]],

            [[ 2.8153,  1.8787, -4.3839, -1.2112],
             [ 0.3728, -2.1131,  0.0921,  0.8305]]])

    >>> A = torch.randn(3, 3)
    >>> torch.einsum('ii->i', A) # diagonal
    tensor([-0.7825,  0.8291, -0.1936])

    >>> A = torch.randn(4, 3, 3)
    >>> torch.einsum('...ii->...i', A) # batch diagonal
    tensor([[-1.0864,  0.7292,  0.0569],
            [-0.9725, -1.0270,  0.6493],
            [ 0.5832, -1.1716, -1.5084],
            [ 0.4041, -1.1690,  0.8570]])

    >>> A = torch.randn(2, 3, 4, 5)
    >>> torch.einsum('...ij->...ji', A).shape # batch permute
    torch.Size([2, 3, 5, 4])
r   r   )ÚlenÚ
isinstancer    Útupler   r   r   r   )ZequationZoperandsr   r   r   r   ƒ   s    Cc             C   sJ   t | tjƒstdjt| ƒƒƒ‚| jƒ s6tj| tjd�S | | k| j	ƒ t
k@ S )aÎ  Returns a new tensor with boolean elements representing if each element is `Finite` or not.

    Arguments:
        tensor (Tensor): A tensor to check

    Returns:
        Tensor: ``A torch.Tensor with dtype torch.bool`` containing a True at each location of finite elements and False otherwise

    Example::

        >>> torch.isfinite(torch.tensor([1, float('inf'), 2, float('-inf'), float('nan')]))
        tensor([True,  False,  True,  False,  False])
    z The argument is not a tensor: {})r   )r3   r   ÚTensorÚ	TypeErrorÚformatÚreprZis_floating_pointZ	ones_likeÚboolÚabsr   )r   r   r   r   r	   Ì   s
    c             C   sX   t | tjƒstdjt| ƒƒƒ‚| jtjtjtj	tj
tjgkrLtj| tjd�S | jƒ tkS )aÍ  Returns a new tensor with boolean elements representing if each element is `+/-INF` or not.

    Arguments:
        tensor (Tensor): A tensor to check

    Returns:
        Tensor: ``A torch.Tensor with dtype torch.bool`` containing a True at each location of `+/-INF` elements and False otherwise

    Example::

        >>> torch.isinf(torch.tensor([1, float('inf'), 2, float('-inf'), float('nan')]))
        tensor([False,  True,  False,  True,  False])
    z The argument is not a tensor: {})r   )r3   r   r5   r6   r7   r8   r   Zuint8Zint8Zint16Zint32Zint64Z
zeros_liker9   r:   r   )r   r   r   r   r
   æ   s
    c              O   sN   |rt dt|ƒd f ƒ‚t| ƒdkr@t| d ttfƒr@| d } tjjj| ƒS )a  Take :math:`N` tensors, each of which can be either scalar or 1-dimensional
vector, and create :math:`N` N-dimensional grids, where the :math:`i` :sup:`th` grid is defined by
expanding the :math:`i` :sup:`th` input over dimensions defined by other inputs.


    Args:
        tensors (list of Tensor): list of scalars or 1 dimensional tensors. Scalars will be
        treated as tensors of size :math:`(1,)` automatically

    Returns:
        seq (sequence of Tensors): If the input has :math:`k` tensors of size
        :math:`(N_1,), (N_2,), \ldots , (N_k,)`, then the output would also have :math:`k` tensors,
        where all tensors are of size :math:`(N_1, N_2, \ldots , N_k)`.

    Example::

        >>> x = torch.tensor([1, 2, 3])
        >>> y = torch.tensor([4, 5, 6])
        >>> grid_x, grid_y = torch.meshgrid(x, y)
        >>> grid_x
        tensor([[1, 1, 1],
                [2, 2, 2],
                [3, 3, 3]])
        >>> grid_y
        tensor([[4, 5, 6],
                [4, 5, 6],
                [4, 5, 6]])
    z2meshgrid() got an unexpected keyword argument '%s'r   r   )	r6   r    r2   r3   r4   r   r   r   r   )r   Úkwargsr   r   r   r   û   s
    ÚreflectFc	             C   sz   |r`| j ƒ }	dgd|	  t| jƒ ƒ }
t|d ƒ}tj| j|
ƒ||f|ƒ} | j| j|	 d… ƒ} tj	j
j| ||||||ƒS )a  Short-time Fourier transform (STFT).

    Ignoring the optional batch dimension, this method computes the following
    expression:

    .. math::
        X[m, \omega] = \sum_{k = 0}^{\text{win\_length-1}}%
                            \text{window}[k]\ \text{input}[m \times \text{hop\_length} + k]\ %
                            \exp\left(- j \frac{2 \pi \cdot \omega k}{\text{win\_length}}\right),

    where :math:`m` is the index of the sliding window, and :math:`\omega` is
    the frequency that :math:`0 \leq \omega < \text{n\_fft}`. When
    :attr:`onesided` is the default value ``True``,

    * :attr:`input` must be either a 1-D time sequence or a 2-D batch of time
      sequences.

    * If :attr:`hop_length` is ``None`` (default), it is treated as equal to
      ``floor(n_fft / 4)``.

    * If :attr:`win_length` is ``None`` (default), it is treated as equal to
      :attr:`n_fft`.

    * :attr:`window` can be a 1-D tensor of size :attr:`win_length`, e.g., from
      :meth:`torch.hann_window`. If :attr:`window` is ``None`` (default), it is
      treated as if having :math:`1` everywhere in the window. If
      :math:`\text{win\_length} < \text{n\_fft}`, :attr:`window` will be padded on
      both sides to length :attr:`n_fft` before being applied.

    * If :attr:`center` is ``True`` (default), :attr:`input` will be padded on
      both sides so that the :math:`t`-th frame is centered at time
      :math:`t \times \text{hop\_length}`. Otherwise, the :math:`t`-th frame
      begins at time  :math:`t \times \text{hop\_length}`.

    * :attr:`pad_mode` determines the padding method used on :attr:`input` when
      :attr:`center` is ``True``. See :meth:`torch.nn.functional.pad` for
      all available options. Default is ``"reflect"``.

    * If :attr:`onesided` is ``True`` (default), only values for :math:`\omega`
      in :math:`\left[0, 1, 2, \dots, \left\lfloor \frac{\text{n\_fft}}{2} \right\rfloor + 1\right]`
      are returned because the real-to-complex Fourier transform satisfies the
      conjugate symmetry, i.e., :math:`X[m, \omega] = X[m, \text{n\_fft} - \omega]^*`.

    * If :attr:`normalized` is ``True`` (default is ``False``), the function
      returns the normalized STFT results, i.e., multiplied by :math:`(\text{frame\_length})^{-0.5}`.

    Returns the real and the imaginary parts together as one tensor of size
    :math:`(* \times N \times T \times 2)`, where :math:`*` is the optional
    batch size of :attr:`input`, :math:`N` is the number of frequencies where
    STFT is applied, :math:`T` is the total number of frames used, and each pair
    in the last dimension represents a complex number as the real part and the
    imaginary part.

    .. warning::
      This function changed signature at version 0.4.1. Calling with the
      previous signature may cause error or return incorrect result.

    Arguments:
        input (Tensor): the input tensor
        n_fft (int): size of Fourier transform
        hop_length (int, optional): the distance between neighboring sliding window
            frames. Default: ``None`` (treated as equal to ``floor(n_fft / 4)``)
        win_length (int, optional): the size of window frame and STFT filter.
            Default: ``None``  (treated as equal to :attr:`n_fft`)
        window (Tensor, optional): the optional window function.
            Default: ``None`` (treated as window of all :math:`1` s)
        center (bool, optional): whether to pad :attr:`input` on both sides so
            that the :math:`t`-th frame is centered at time :math:`t \times \text{hop\_length}`.
            Default: ``True``
        pad_mode (string, optional): controls the padding method used when
            :attr:`center` is ``True``. Default: ``"reflect"``
        normalized (bool, optional): controls whether to return the normalized STFT results
             Default: ``False``
        onesided (bool, optional): controls whether to return half of results to
            avoid redundancy Default: ``True``

    Returns:
        Tensor: A tensor containing the STFT result with shape described above

    r   é   r   N)r   r    r&   ÚintÚFÚpadÚviewr)   r   r   r   r   )ÚinputZn_fftZ
hop_lengthZ
win_lengthZwindowÚcenterZpad_modeÚ
normalizedZonesidedZ
signal_dimZextended_shaper@   r   r   r   r      s    Uc             C   sr   |dk	r(t jjj| ||||d�\}}}nt j| |||d�\}}}|rR|rR|||fS |r^||fS |rj||fS |S dS )aˆ  Returns the unique elements of the input tensor.

    Arguments:
        input (Tensor): the input tensor
        sorted (bool): Whether to sort the unique elements in ascending order
            before returning as output.
        return_inverse (bool): Whether to also return the indices for where
            elements in the original input ended up in the returned unique list.
        return_counts (bool): Whether to also return the counts for each unique
            element.
        dim (int): the dimension to apply unique. If ``None``, the unique of the
            flattened input is returned. default: ``None``

    Returns:
        (Tensor, Tensor (optional), Tensor (optional)): A tensor or a tuple of tensors containing

            - **output** (*Tensor*): the output list of unique scalar elements.
            - **inverse_indices** (*Tensor*): (optional) if
              :attr:`return_inverse` is True, there will be an additional
              returned tensor (same shape as input) representing the indices
              for where elements in the original input map to in the output;
              otherwise, this function will only return a single tensor.
            - **counts** (*Tensor*): (optional) if
              :attr:`return_counts` is True, there will be an additional
              returned tensor (same shape as output or output.size(dim),
              if dim was specified) representing the number of occurrences
              for each unique value or tensor.

    Example::

        >>> output = torch.unique(torch.tensor([1, 3, 2, 3], dtype=torch.long))
        >>> output
        tensor([ 2,  3,  1])

        >>> output, inverse_indices = torch.unique(
                torch.tensor([1, 3, 2, 3], dtype=torch.long), sorted=True, return_inverse=True)
        >>> output
        tensor([ 1,  2,  3])
        >>> inverse_indices
        tensor([ 0,  2,  1,  2])

        >>> output, inverse_indices = torch.unique(
                torch.tensor([[1, 3], [2, 3]], dtype=torch.long), sorted=True, return_inverse=True)
        >>> output
        tensor([ 1,  2,  3])
        >>> inverse_indices
        tensor([[ 0,  2],
                [ 1,  2]])

    N)ÚsortedÚreturn_inverseÚreturn_counts)r   r   r   Ú
unique_dimZ_unique2)rB   rE   rF   rG   r   ÚoutputÚinverse_indicesÚcountsr   r   r   r   �  s&    3
c             C   sJ   t jjj| |||d�\}}}|r.|r.|||fS |r:||fS |rF||fS |S )a’  Eliminates all but the first element from every consecutive group of equivalent elements.

    .. note:: This function is different from :func:`torch.unique` in the sense that this function
        only eliminates consecutive duplicate values. This semantics is similar to `std::unique`
        in C++.

    Arguments:
        input (Tensor): the input tensor
        return_inverse (bool): Whether to also return the indices for where
            elements in the original input ended up in the returned unique list.
        return_counts (bool): Whether to also return the counts for each unique
            element.
        dim (int): the dimension to apply unique. If ``None``, the unique of the
            flattened input is returned. default: ``None``

    Returns:
        (Tensor, Tensor (optional), Tensor (optional)): A tensor or a tuple of tensors containing

            - **output** (*Tensor*): the output list of unique scalar elements.
            - **inverse_indices** (*Tensor*): (optional) if
              :attr:`return_inverse` is True, there will be an additional
              returned tensor (same shape as input) representing the indices
              for where elements in the original input map to in the output;
              otherwise, this function will only return a single tensor.
            - **counts** (*Tensor*): (optional) if
              :attr:`return_counts` is True, there will be an additional
              returned tensor (same shape as output or output.size(dim),
              if dim was specified) representing the number of occurrences
              for each unique value or tensor.

    Example::

        >>> x = torch.tensor([1, 1, 2, 2, 3, 1, 1, 2])
        >>> output = torch.unique_consecutive(x)
        >>> output
        tensor([1, 2, 3, 1, 2])

        >>> output, inverse_indices = torch.unique_consecutive(x, return_inverse=True)
        >>> output
        tensor([1, 2, 3, 1, 2])
        >>> inverse_indices
        tensor([0, 0, 1, 1, 2, 3, 3, 4])

        >>> output, counts = torch.unique_consecutive(x, return_counts=True)
        >>> output
        tensor([1, 2, 3, 1, 2])
        >>> counts
        tensor([2, 2, 1, 2, 1])
    )rF   rG   r   )r   r   r   r   )rB   rF   rG   r   rI   rJ   rK   r   r   r   r   Í  s    2
r   c             C   st   t |ttfƒs&t |tjƒr0|jƒ dkr0|\}}n0t |tjƒrD|jƒ }tt| dƒƒ}tt|ƒƒ}tjj	j
| |||ƒS )aÛ  Returns a contraction of a and b over multiple dimensions.

    :attr:`tensordot` implements a generalized matrix product.

    Args:
      a (Tensor): Left tensor to contract
      b (Tensor): Right tensor to contract
      dims (int or tuple of two lists of integers): number of dimensions to
         contract or explicit lists of dimensions for :attr:`a` and
         :attr:`b` respectively

    When called with an integer argument :attr:`dims` = :math:`d`, and the number of
    dimensions of :attr:`a` and :attr:`b` is :math:`m` and :math:`n`, respectively,
    it computes

    .. math::
        r_{i_0,...,i_{m-d}, i_d,...,i_n}
          = \sum_{k_0,...,k_{d-1}} a_{i_0,...,i_{m-d},k_0,...,k_{d-1}} \times b_{k_0,...,k_{d-1}, i_d,...,i_n}.

    When called with :attr:`dims` of the list form, the given dimensions will be contracted
    in place of the last :math:`d` of :attr:`a` and the first :math:`d` of :math:`b`. The sizes
    in these dimensions must match, but :attr:`tensordot` will deal with broadcasted
    dimensions.

    Examples::

        >>> a = torch.arange(60.).reshape(3, 4, 5)
        >>> b = torch.arange(24.).reshape(4, 3, 2)
        >>> torch.tensordot(a, b, dims=([1, 0], [0, 1]))
        tensor([[4400., 4730.],
                [4532., 4874.],
                [4664., 5018.],
                [4796., 5162.],
                [4928., 5306.]])

        >>> a = torch.randn(3, 4, 5, device='cuda')
        >>> b = torch.randn(4, 5, 6, device='cuda')
        >>> c = torch.tensordot(a, b, dims=2).cpu()
        tensor([[ 8.3504, -2.5436,  6.2922,  2.7556, -1.0732,  3.2741],
                [ 3.3161,  0.0704,  5.0187, -0.4079, -4.3126,  4.8744],
                [ 0.8223,  3.9445,  3.2168, -0.2400,  3.4117,  1.7780]])

    r   r   )r3   r    r4   r   r5   ZnumelÚitemr!   r   r   r   )ÚaÚbZdimsZdims_aZdims_br   r   r   r   
  s    ,
c              G   s   t jjj| ƒS )aL  Do cartesian product of the given sequence of tensors. The behavior is similar to
    python's `itertools.product`.

    Arguments:
        *tensors: any number of 1 dimensional tensors.

    Returns:
        Tensor: A tensor equivalent to converting all the input tensors into lists,
            do `itertools.product` on these lists, and finally convert the resulting list
            into tensor.

    Example::

        >>> a = [1, 2, 3]
        >>> b = [4, 5]
        >>> list(itertools.product(a, b))
        [(1, 4), (1, 5), (2, 4), (2, 5), (3, 4), (3, 5)]
        >>> tensor_a = torch.tensor(a)
        >>> tensor_b = torch.tensor(b)
        >>> torch.cartesian_prod(tensor_a, tensor_b)
        tensor([[1, 4],
                [1, 5],
                [2, 4],
                [2, 5],
                [3, 4],
                [3, 5]])
    )r   r   r   r   )r   r   r   r   r   A  s    Úfroc             C   s²  | j ƒ }|dkrN|dkrN|dkrN|dkr6tjjj| ƒS |dkrNtjjj| |ƒS |dkr¬|dk	rftdƒ‚|dkrztt|ƒƒ}|dkr–tjjj| ||d�S tjjj| |||d�S |dk�r|dk	rÆtdƒ‚|dk� rþ|dkrêtjjj	| |d�S tjjj	| ||d�S tjjj	| |||d�S |dk�r*tt|ƒƒ}|dk�rT|dk�rTtjjj| |||d�S |dk�rvtjjj| ||||d�S |dk�r˜tjjj| ||||d�S tjjj| |||||d	�S )
až
  Returns the matrix norm or vector norm of a given tensor.

    Args:
        input (Tensor): the input tensor
        p (int, float, inf, -inf, 'fro', 'nuc', optional): the order of norm. Default: ``'fro'``
            The following norms can be calculated:

            =====  ============================  ==========================
            ord    matrix norm                   vector norm
            =====  ============================  ==========================
            None   Frobenius norm                2-norm
            'fro'  Frobenius norm                --
            'nuc'  nuclear norm                  --
            Other  as vec norm when dim is None  sum(abs(x)**ord)**(1./ord)
            =====  ============================  ==========================

        dim (int, 2-tuple of ints, 2-list of ints, optional): If it is an int,
            vector norm will be calculated, if it is 2-tuple of ints, matrix norm
            will be calculated. If the value is None, matrix norm will be calculated
            when the input tensor only has two dimensions, vector norm will be
            calculated when the input tensor only has one dimension. If the input
            tensor has more than two dimensions, the vector norm will be applied to
            last dimension.
        keepdim (bool, optional): whether the output tensors have :attr:`dim`
            retained or not. Ignored if :attr:`dim` = ``None`` and
            :attr:`out` = ``None``. Default: ``False``
        out (Tensor, optional): the output tensor. Ignored if
            :attr:`dim` = ``None`` and :attr:`out` = ``None``.
        dtype (:class:`torch.dtype`, optional): the desired data type of
            returned tensor. If specified, the input tensor is casted to
            :attr:'dtype' while performing the operation. Default: None.


    Example::

        >>> import torch
        >>> a = torch.arange(9, dtype= torch.float) - 4
        >>> b = a.reshape((3, 3))
        >>> torch.norm(a)
        tensor(7.7460)
        >>> torch.norm(b)
        tensor(7.7460)
        >>> torch.norm(a, float('inf'))
        tensor(4.)
        >>> torch.norm(b, float('inf'))
        tensor(4.)
        >>> c = torch.tensor([[ 1, 2, 3],[-1, 1, 4]] , dtype= torch.float)
        >>> torch.norm(c, dim=0)
        tensor([1.4142, 2.2361, 5.0000])
        >>> torch.norm(c, dim=1)
        tensor([3.7417, 4.2426])
        >>> torch.norm(c, p=1, dim=1)
        tensor([6., 6.])
        >>> d = torch.arange(8, dtype= torch.float).reshape(2,2,2)
        >>> torch.norm(d, dim=(1,2))
        tensor([ 3.7417, 11.2250])
        >>> torch.norm(d[0, :, :]), torch.norm(d[1, :, :])
        (tensor(3.7417), tensor(11.2250))
    NrO   Znucz1dtype argument is not supported in frobenius norm)Úkeepdim)rP   Úoutz/dtype argument is not supported in nuclear norm)rP   r   )rP   r   rQ   )
r   r   r   r   Zfrobenius_normr   Ú
ValueErrorr4   r!   Znuclear_norm)rB   Úpr   rP   rQ   r   Úndimr   r   r   r   `  s>    <




c              G   s   t jjj| ƒS )aR  Returns the matrix product of the :math:`N` 2-D tensors. This product is efficiently computed
    using the matrix chain order algorithm which selects the order in which incurs the lowest cost in terms
    of arithmetic operations (`[CLRS]`_). Note that since this is a function to compute the product, :math:`N`
    needs to be greater than or equal to 2; if equal to 2 then a trivial matrix-matrix product is returned.
    If :math:`N` is 1, then this is a no-op - the original matrix is returned as is.


    Args:
        matrices (Tensors...): a sequence of 2 or more 2-D tensors whose product is to be determined.


    Returns:
        Tensor: if the :math:`i^{th}` tensor was of dimensions :math:`p_{i} \times p_{i + 1}`, then the product
        would be of dimensions :math:`p_{1} \times p_{N + 1}`.

    Example::

        >>> a = torch.randn(3, 4)
        >>> b = torch.randn(4, 5)
        >>> c = torch.randn(5, 6)
        >>> d = torch.randn(6, 7)
        >>> torch.chain_matmul(a, b, c, d)
        tensor([[ -2.3375,  -3.9790,  -4.1119,  -6.6577,   9.5609, -11.5095,  -3.2614],
                [ 21.4038,   3.3378,  -8.4982,  -5.2457, -10.2561,  -2.4684,   2.7163],
                [ -0.9647,  -5.8917,  -2.3213,  -5.2284,  12.8615, -12.2816,  -2.5095]])

    .. _`[CLRS]`: https://mitpress.mit.edu/books/introduction-algorithms-third-edition
    )r   r   r   r   )Zmatricesr   r   r   r   Á  s    c                s¤   t j| || d�‰ˆ dk	rˆtˆ ttfƒs<tdjtˆ ƒjƒƒ‚t	ˆ ƒt
|ƒ dkrltdjdt
|ƒ t	ˆ ƒƒƒ‚‡ ‡fdd„tt	ˆ ƒƒD ƒS |r�ˆS ˆd ˆd	 fS dS )
aæ
  Computes the LU factorization of a square matrix or batches of square matrices
    :attr:`A`. Returns a tuple containing the LU factorization and pivots of :attr:`A`.
    Pivoting is done if :attr:`pivot` is set to ``True``.

    .. note::
        The pivots returned by the function are 1-indexed. If :attr:`pivot` is ``False``,
        then the returned pivots is a tensor filled with zeros of the appropriate size.

    .. note::
        LU factorization with :attr:`pivot` = ``False`` is not available for CPU, and attempting
        to do so will throw an error. However, LU factorization with :attr:`pivot` = ``False`` is
        available for CUDA.

    .. note::
        This function does not check if the factorization was successful or not if
        :attr:`get_infos` is ``True`` since the status of the factorization is present in the
        third element of the return tuple.

    Arguments:
        A (Tensor): the tensor to factor of size :math:`(*, m, m)`
        pivot (bool, optional): controls whether pivoting is done. Default: ``True``
        get_infos (bool, optional): if set to ``True``, returns an info IntTensor.
                                    Default: ``False``
        out (tuple, optional): optional output tuple. If :attr:`get_infos` is ``True``,
                               then the elements in the tuple are Tensor, IntTensor,
                               and IntTensor. If :attr:`get_infos` is ``False``, then the
                               elements in the tuple are Tensor, IntTensor. Default: ``None``

    Returns:
        (Tensor, IntTensor, IntTensor (optional)): A tuple of tensors containing

            - **factorization** (*Tensor*): the factorization of size :math:`(*, m, m)`

            - **pivots** (*IntTensor*): the pivots of size :math:`(*, m)`

            - **infos** (*IntTensor*, *optional*): if :attr:`get_infos` is ``True``, this is a tensor of
              size :math:`(*)` where non-zero values indicate whether factorization for the matrix or
              each minibatch has succeeded or failed

    Example::

        >>> A = torch.randn(2, 3, 3)
        >>> A_LU, pivots = torch.lu(A)
        >>> A_LU
        tensor([[[ 1.3506,  2.5558, -0.0816],
                 [ 0.1684,  1.1551,  0.1940],
                 [ 0.1193,  0.6189, -0.5497]],

                [[ 0.4526,  1.2526, -0.3285],
                 [-0.7988,  0.7175, -0.9701],
                 [ 0.2634, -0.9255, -0.3459]]])
        >>> pivots
        tensor([[ 3,  3,  3],
                [ 3,  3,  3]], dtype=torch.int32)
        >>> A_LU, pivots, info = torch.lu(A, get_infos=True)
        >>> if info.nonzero().size(0) == 0:
        ...   print('LU factorization succeeded for all samples!')
        LU factorization succeeded for all samples!
    )ÚpivotZcheck_errorsNz/argument 'out' must be tuple of Tensors, not {}r   z(expected tuple of {} elements but got {}c             3   s*   | ]"}ˆ | j ˆ| ƒjˆ| ƒV  qd S )N)Z
resize_as_Zcopy_)Ú.0Úi)rQ   Úresultr   r   ú	<genexpr>&  s    zlu.<locals>.<genexpr>r   r   )r   Z_lu_with_infor3   r4   r    r6   r7   ÚtypeÚ__name__r2   r>   r!   )ÚArU   Z	get_infosrQ   r   )rQ   rX   r   r   á  s    =c              G   s   t dƒ‚d S )Nz$`align_tensors` not yet implemented.)ÚRuntimeError)r   r   r   r   r   -  s    )r   )TT)NNNTr<   FT)TFFN)FFN)r   )rO   NFNN)TFN)r   Ztorch.nn.functionalÚnnZ
functionalr?   Z
torch._sixr   Ú	itertoolsr   Ú__all__r   r   r   r   r	   r
   r   r   rH   r   r   r   r   r   r   r   r   r   r   r   r   Ú<module>   sJ   

3I% 
]
L
=
7
a 
L