r"""Element-wise addition. Calculates the sum for each element :math:`x_i` of the input tensor :math:`x` with the respective element :math:`y

! * \brief Get the underlying megcore computing handle. */

r"""Get a new TracedModule, which eliminates ``GetAttr`` and has no hierarchy. Retruns: A new :class:`TracedModule`.

! * \brief whether var shape could be statically inferred */

:param cases: the list which have dict element, the list length should be 2 for dynamic shape test. and the dict should have input,

Natural logorithm. \param arg function argument \return logarithm of \a arg to base e template<typename T> typename enable<expr,T>::type log(T arg) {

r"""Calculates the sum of tensor elements over a given axis (or axes). Args: inp: input tensor. Should have a numeric data type.

get the buffer of the tensor

* \brief The collapse_contiguous function will convert a contiguous image like * tensor layout into a 2-dimensional layout, shape[0] = height of the

r"""Gets number of devices installed on this system. Args: device_type: device type, one of 'gpu' or 'cpu'

r"""Traces module ``mod`` and returns corresponding :class:`TracedModule`. Args: mod(:attr:`.module.Module`): the module will be converte

r"""Copies tensor to another device. Args: inp: input tensor. device: destination device. Examples: >>> import nump

! * \brief get the interval whose address is greater than newly * allocated one, or nullptr if the new interval is at end */

r"""Sets training mode of all the modules within this module (including itself) to ``mode``. This effectively sets the ``training`` attributes

Used for initialization

r"""Flattens a pytree into a list of values and a :class:`TreeDef` that can be used to reconstruct the pytree.

! * \brief previous dev ptr before deallocating dev_tensor; used for * testing and debugging */

r"""Copy self :class:`~.Tensor` to specified device. See :func:`~.copy`