blob: 31258631a25480994e96043e524f7eab72dcef01 [file] [log] [blame]
Symbol
====================================================
Symbolic programming with computation graphs
------------------------------------------------------
Get symbol attributes
-----------------------
======================================================== ==================================================================================================
:doc:`arguments <./arguments>` Get the arguments of symbol
:doc:`children <./children>` Gets a new grouped symbol whose output contains inputs to output nodes of the original symbol
:doc:`graph.viz <./graph.viz>` Convert symbol to Graphviz or visNetwork visualisation
:doc:`internals <./internals>` Get a symbol that contains all the internals
:doc:`is.mx.symbol <./is.mx.symbol>` Judge if an object is mx.symbol
:doc:`mx.symbol.infer.shape <./mx.symbol.infer.shape>` Inference the shape of arguments, outputs, and auxiliary states
:doc:`mx.symbol.shape_array <./mx.symbol.shape_array>` Returns a 1D int64 array containing the shape of data
:doc:`mx.symbol.size_array <./mx.symbol.size_array>` Returns a 1D int64 array containing the size of data
:doc:`outputs <./outputs>` Get the outputs of a symbol
======================================================== ==================================================================================================
.. toctree::
:titlesonly:
:hidden:
arguments
children
graph.viz
internals
is.mx.symbol
mx.symbol.infer.shape
mx.symbol.shape_array
mx.symbol.size_array
outputs
Create symbols
------------------
====================================================== ===========================================================================================
:doc:`mx.symbol.Group <./mx.symbol.Group>` Create a symbol that groups symbols together
:doc:`mx.symbol.GroupNorm <./mx.symbol.GroupNorm>` Group normalization
:doc:`mx.symbol.Variable <./mx.symbol.Variable>` Create a symbolic variable with specified name
:doc:`mx.symbol.one_hot <./mx.symbol.one_hot>` Returns a one-hot array
:doc:`mx.symbol.ones_like <./mx.symbol.ones_like>` Return an array of ones with the same shape and type as the input array
:doc:`mx.symbol.zeros_like <./mx.symbol.zeros_like>` Return an array of zeros with the same shape, type and storage type as the input array
====================================================== ===========================================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.Group
mx.symbol.GroupNorm
mx.symbol.Variable
mx.symbol.one_hot
mx.symbol.ones_like
mx.symbol.zeros_like
Manipulation of symbols
------------------------
Conversion
^^^^^^^^^^^^
========================================================== ==================================================
:doc:`mx.apply <./mx.apply>` Apply symbol to the inputs
:doc:`mx.symbol.cast <./mx.symbol.cast>` Casts all elements of the input to a new type
:doc:`mx.symbol.cast_storage <./mx.symbol.cast_storage>` Casts tensor storage type to the new type
:doc:`mx.symbol.load <./mx.symbol.load>` Load an mx.symbol object
:doc:`mx.symbol.load.json <./mx.symbol.load.json>` Load an mx.symbol object from a json string
:doc:`mx.symbol.save <./mx.symbol.save>` Save an mx.symbol object
========================================================== ==================================================
.. toctree::
:titlesonly:
:hidden:
mx.apply
mx.symbol.cast
mx.symbol.cast_storage
mx.symbol.load
mx.symbol.load.json
mx.symbol.save
Reshaping
^^^^^^^^^^^^^
========================================================== ======================================================================================================
:doc:`mx.symbol.Concat <./mx.symbol.Concat>` Perform an feature concat on channel dim (dim 1) over all the inputs
:doc:`mx.symbol.Reshape <./mx.symbol.Reshape>` Reshapes the input array
:doc:`mx.symbol.flatten <./mx.symbol.flatten>` Flattens the input array into a 2-D array by collapsing the higher dimensions
:doc:`mx.symbol.reshape_like <./mx.symbol.reshape_like>` Reshape some or all dimensions of `lhs` to have the same shape as some or all dimensions of `rhs`
:doc:`mx.symbol.scatter_nd <./mx.symbol.scatter_nd>` Scatters data into a new tensor according to indices
:doc:`mx.symbol.split <./mx.symbol.split>` Splits an array along a particular axis into multiple sub-arrays
:doc:`mx.symbol.squeeze <./mx.symbol.squeeze>` Remove single-dimensional entries from the shape of an array
:doc:`mx.symbol.stack <./mx.symbol.stack>` Join a sequence of arrays along a new axis
========================================================== ======================================================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.Concat
mx.symbol.Reshape
mx.symbol.flatten
mx.symbol.reshape_like
mx.symbol.scatter_nd
mx.symbol.split
mx.symbol.squeeze
mx.symbol.stack
Expanding elements
^^^^^^^^^^^^^^^^^^^^^
================================================================================ ========================================================================================================================================================================================
:doc:`mx.symbol.GridGenerator <./mx.symbol.GridGenerator>` Generates 2D sampling grid for bilinear sampling
:doc:`mx.symbol.Pad <./mx.symbol.Pad>` Pads an input array with a constant or edge values of the array
:doc:`mx.symbol.UpSampling <./mx.symbol.UpSampling>` Upsamples the given input data
:doc:`mx.symbol.broadcast_add <./mx.symbol.broadcast_add>` Returns element-wise sum of the input arrays with broadcasting
:doc:`mx.symbol.broadcast_axes <./mx.symbol.broadcast_axes>` Broadcasts the input array over particular axes
:doc:`mx.symbol.broadcast_axis <./mx.symbol.broadcast_axis>` Broadcasts the input array over particular axes
:doc:`mx.symbol.broadcast_div <./mx.symbol.broadcast_div>` Returns element-wise division of the input arrays with broadcasting
:doc:`mx.symbol.broadcast_equal <./mx.symbol.broadcast_equal>` Returns the result of element-wise **equal to** (==) comparison operation with broadcasting
:doc:`mx.symbol.broadcast_greater <./mx.symbol.broadcast_greater>` Returns the result of element-wise **greater than** (>) comparison operation with broadcasting
:doc:`mx.symbol.broadcast_greater_equal <./mx.symbol.broadcast_greater_equal>` Returns the result of element-wise **greater than or equal to** (>=) comparison operation with broadcasting
:doc:`mx.symbol.broadcast_hypot <./mx.symbol.broadcast_hypot>` Returns the hypotenuse of a right angled triangle, given its "legs" with broadcasting
:doc:`mx.symbol.broadcast_lesser <./mx.symbol.broadcast_lesser>` Returns the result of element-wise **lesser than** (<) comparison operation with broadcasting
:doc:`mx.symbol.broadcast_lesser_equal <./mx.symbol.broadcast_lesser_equal>` Returns the result of element-wise **lesser than or equal to** (<=) comparison operation with broadcasting
:doc:`mx.symbol.broadcast_like <./mx.symbol.broadcast_like>` Broadcasts lhs to have the same shape as rhs
:doc:`mx.symbol.broadcast_logical_and <./mx.symbol.broadcast_logical_and>` Returns the result of element-wise **logical and** with broadcasting
:doc:`mx.symbol.broadcast_logical_or <./mx.symbol.broadcast_logical_or>` Returns the result of element-wise **logical or** with broadcasting
:doc:`mx.symbol.broadcast_logical_xor <./mx.symbol.broadcast_logical_xor>` Returns the result of element-wise **logical xor** with broadcasting
:doc:`mx.symbol.broadcast_maximum <./mx.symbol.broadcast_maximum>` Returns element-wise maximum of the input arrays with broadcasting
:doc:`mx.symbol.broadcast_minimum <./mx.symbol.broadcast_minimum>` Returns element-wise minimum of the input arrays with broadcasting
:doc:`mx.symbol.broadcast_minus <./mx.symbol.broadcast_minus>` Returns element-wise difference of the input arrays with broadcasting
:doc:`mx.symbol.broadcast_mod <./mx.symbol.broadcast_mod>` Returns element-wise modulo of the input arrays with broadcasting
:doc:`mx.symbol.broadcast_mul <./mx.symbol.broadcast_mul>` Returns element-wise product of the input arrays with broadcasting
:doc:`mx.symbol.broadcast_not_equal <./mx.symbol.broadcast_not_equal>` Returns the result of element-wise **not equal to** (!=) comparison operation with broadcasting
:doc:`mx.symbol.broadcast_plus <./mx.symbol.broadcast_plus>` Returns element-wise sum of the input arrays with broadcasting
:doc:`mx.symbol.broadcast_power <./mx.symbol.broadcast_power>` Returns result of first array elements raised to powers from second array, element-wise with broadcasting
:doc:`mx.symbol.broadcast_sub <./mx.symbol.broadcast_sub>` Returns element-wise difference of the input arrays with broadcasting
:doc:`mx.symbol.broadcast_to <./mx.symbol.broadcast_to>` Broadcasts the input array to a new shape
:doc:`mx.symbol.expand_dims <./mx.symbol.expand_dims>` Inserts a new axis of size 1 into the array shape For example, given ``x`` with shape ``(2,3,4)``, then ``expand_dims(x, axis=1)`` will return a new array with shape ``(2,1,3,4)``
:doc:`mx.symbol.repeat <./mx.symbol.repeat>` Repeats elements of an array
:doc:`mx.symbol.tile <./mx.symbol.tile>` Repeats the whole array multiple times
================================================================================ ========================================================================================================================================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.GridGenerator
mx.symbol.Pad
mx.symbol.UpSampling
mx.symbol.broadcast_add
mx.symbol.broadcast_axes
mx.symbol.broadcast_axis
mx.symbol.broadcast_div
mx.symbol.broadcast_equal
mx.symbol.broadcast_greater
mx.symbol.broadcast_greater_equal
mx.symbol.broadcast_hypot
mx.symbol.broadcast_lesser
mx.symbol.broadcast_lesser_equal
mx.symbol.broadcast_like
mx.symbol.broadcast_logical_and
mx.symbol.broadcast_logical_or
mx.symbol.broadcast_logical_xor
mx.symbol.broadcast_maximum
mx.symbol.broadcast_minimum
mx.symbol.broadcast_minus
mx.symbol.broadcast_mod
mx.symbol.broadcast_mul
mx.symbol.broadcast_not_equal
mx.symbol.broadcast_plus
mx.symbol.broadcast_power
mx.symbol.broadcast_sub
mx.symbol.broadcast_to
mx.symbol.expand_dims
mx.symbol.repeat
mx.symbol.tile
Rearranging elements
^^^^^^^^^^^^^^^^^^^^^^
================================================================ =================================================================================
:doc:`mx.symbol.SequenceReverse <./mx.symbol.SequenceReverse>` Reverses the elements of each sequence
:doc:`mx.symbol.SwapAxis <./mx.symbol.SwapAxis>` Interchanges two axes of an array
:doc:`mx.symbol.depth_to_space <./mx.symbol.depth_to_space>` Rearranges(permutes) data from depth into blocks of spatial data
:doc:`mx.symbol.flip <./mx.symbol.flip>` Reverses the order of elements along given axis while preserving array shape
:doc:`mx.symbol.reverse <./mx.symbol.reverse>` Reverses the order of elements along given axis while preserving array shape
:doc:`mx.symbol.shuffle <./mx.symbol.shuffle>` Randomly shuffle the elements
:doc:`mx.symbol.space_to_depth <./mx.symbol.space_to_depth>` Rearranges(permutes) blocks of spatial data into depth
:doc:`mx.symbol.swapaxes <./mx.symbol.swapaxes>` Interchanges two axes of an array
:doc:`mx.symbol.transpose <./mx.symbol.transpose>` Permutes the dimensions of an array
================================================================ =================================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.SequenceReverse
mx.symbol.SwapAxis
mx.symbol.depth_to_space
mx.symbol.flip
mx.symbol.reverse
mx.symbol.shuffle
mx.symbol.space_to_depth
mx.symbol.swapaxes
mx.symbol.transpose
Sorting and searching
^^^^^^^^^^^^^^^^^^^^^^^
============================================================== ====================================================================================================
:doc:`mx.symbol.argmax <./mx.symbol.argmax>` Returns indices of the maximum values along an axis
:doc:`mx.symbol.argmax_channel <./mx.symbol.argmax_channel>` Returns argmax indices of each channel from the input array
:doc:`mx.symbol.argmin <./mx.symbol.argmin>` Returns indices of the minimum values along an axis
:doc:`mx.symbol.argsort <./mx.symbol.argsort>` Returns the indices that would sort an input array along the given axis
:doc:`mx.symbol.sort <./mx.symbol.sort>` Returns a sorted copy of an input array along the given axis
:doc:`mx.symbol.topk <./mx.symbol.topk>` Returns the indices of the top *k* elements in an input array along the given axis (by default)
:doc:`mx.symbol.where <./mx.symbol.where>` Return the elements, either from x or y, depending on the condition
============================================================== ====================================================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.argmax
mx.symbol.argmax_channel
mx.symbol.argmin
mx.symbol.argsort
mx.symbol.sort
mx.symbol.topk
mx.symbol.where
Indexing
^^^^^^^^^^^^
============================================================================ =============================================================================================================================================
:doc:`mx.symbol.Crop <./mx.symbol.Crop>`
:doc:`mx.symbol.SequenceLast <./mx.symbol.SequenceLast>` Takes the last element of a sequence
:doc:`mx.symbol.SequenceMask <./mx.symbol.SequenceMask>` Sets all elements outside the sequence to a constant value
:doc:`mx.symbol.SliceChannel <./mx.symbol.SliceChannel>` Splits an array along a particular axis into multiple sub-arrays
:doc:`mx.symbol.batch_take <./mx.symbol.batch_take>` Takes elements from a data batch
:doc:`mx.symbol.choose_element_0index <./mx.symbol.choose_element_0index>` Picks elements from an input array according to the input indices along the given axis
:doc:`mx.symbol.diag <./mx.symbol.diag>` Extracts a diagonal or constructs a diagonal array
:doc:`mx.symbol.fill_element_0index <./mx.symbol.fill_element_0index>` Fill one element of each line(row for python, column for R/Julia) in lhs according to index indicated by rhs and values indicated by mhs
:doc:`mx.symbol.gather_nd <./mx.symbol.gather_nd>` Gather elements or slices from `data` and store to a tensor whose shape is defined by `indices`
:doc:`mx.symbol.pick <./mx.symbol.pick>` Picks elements from an input array according to the input indices along the given axis
:doc:`mx.symbol.ravel_multi_index <./mx.symbol.ravel_multi_index>` Converts a batch of index arrays into an array of flat indices
:doc:`mx.symbol.slice <./mx.symbol.slice>` Slices a region of the array
:doc:`mx.symbol.slice_axis <./mx.symbol.slice_axis>` Slices along a given axis
:doc:`mx.symbol.slice_like <./mx.symbol.slice_like>` Slices a region of the array like the shape of another array
:doc:`mx.symbol.take <./mx.symbol.take>` Takes elements from an input array along the given axis
:doc:`mx.symbol.unravel_index <./mx.symbol.unravel_index>` Converts an array of flat indices into a batch of index arrays
============================================================================ =============================================================================================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.Crop
mx.symbol.SequenceLast
mx.symbol.SequenceMask
mx.symbol.SliceChannel
mx.symbol.batch_take
mx.symbol.choose_element_0index
mx.symbol.diag
mx.symbol.fill_element_0index
mx.symbol.gather_nd
mx.symbol.pick
mx.symbol.ravel_multi_index
mx.symbol.slice
mx.symbol.slice_axis
mx.symbol.slice_like
mx.symbol.take
mx.symbol.unravel_index
Mathematical operations on symbols
-----------------------------------
Arithmetic
^^^^^^^^^^^^^
================================================================== ==============================================================================================================================
:doc:`mx.symbol.ElementWiseSum <./mx.symbol.ElementWiseSum>` Adds all input arguments element-wise
:doc:`mx.symbol.abs <./mx.symbol.abs>` Returns element-wise absolute value of the input
:doc:`mx.symbol.add_n <./mx.symbol.add_n>` Adds all input arguments element-wise
:doc:`mx.symbol.all_finite <./mx.symbol.all_finite>` Check if all the float numbers in the array are finite (used for AMP)
:doc:`mx.symbol.cbrt <./mx.symbol.cbrt>` Returns element-wise cube-root value of the input
:doc:`mx.symbol.cumsum <./mx.symbol.cumsum>` Return the cumulative sum of the elements along a given axis
:doc:`mx.symbol.elemwise_add <./mx.symbol.elemwise_add>` Adds arguments element-wise
:doc:`mx.symbol.elemwise_div <./mx.symbol.elemwise_div>` Divides arguments element-wise
:doc:`mx.symbol.elemwise_mul <./mx.symbol.elemwise_mul>` Multiplies arguments element-wise
:doc:`mx.symbol.elemwise_sub <./mx.symbol.elemwise_sub>` Subtracts arguments element-wise
:doc:`mx.symbol.erf <./mx.symbol.erf>` Returns element-wise gauss error function of the input
:doc:`mx.symbol.erfinv <./mx.symbol.erfinv>` Returns element-wise inverse gauss error function of the input
:doc:`mx.symbol.exp <./mx.symbol.exp>` Returns element-wise exponential value of the input
:doc:`mx.symbol.expm1 <./mx.symbol.expm1>` Returns ``exp(x) - 1`` computed element-wise on the input
:doc:`mx.symbol.gamma <./mx.symbol.gamma>` Returns the gamma function (extension of the factorial function \ to the reals), computed element-wise on the input array
:doc:`mx.symbol.gammaln <./mx.symbol.gammaln>` Returns element-wise log of the absolute value of the gamma function \ of the input
:doc:`mx.symbol.identity <./mx.symbol.identity>` Returns a copy of the input
:doc:`mx.symbol.log <./mx.symbol.log>` Returns element-wise Natural logarithmic value of the input
:doc:`mx.symbol.log10 <./mx.symbol.log10>` Returns element-wise Base-10 logarithmic value of the input
:doc:`mx.symbol.log1p <./mx.symbol.log1p>` Returns element-wise ``log(1 + x)`` value of the input
:doc:`mx.symbol.log2 <./mx.symbol.log2>` Returns element-wise Base-2 logarithmic value of the input
:doc:`mx.symbol.logical_not <./mx.symbol.logical_not>` Returns the result of logical NOT (!) function
:doc:`mx.symbol.moments <./mx.symbol.moments>` Calculate the mean and variance of `data`
:doc:`mx.symbol.multi_all_finite <./mx.symbol.multi_all_finite>` Check if all the float numbers in all the arrays are finite (used for AMP)
:doc:`mx.symbol.multi_lars <./mx.symbol.multi_lars>` Compute the LARS coefficients of multiple weights and grads from their sums of square"
:doc:`mx.symbol.multi_sum_sq <./mx.symbol.multi_sum_sq>` Compute the sums of squares of multiple arrays
:doc:`mx.symbol.negative <./mx.symbol.negative>` Numerical negative of the argument, element-wise
:doc:`mx.symbol.rcbrt <./mx.symbol.rcbrt>` Returns element-wise inverse cube-root value of the input
:doc:`mx.symbol.reciprocal <./mx.symbol.reciprocal>` Returns the reciprocal of the argument, element-wise
:doc:`mx.symbol.reset_arrays <./mx.symbol.reset_arrays>` Set to zero multiple arrays
:doc:`mx.symbol.rsqrt <./mx.symbol.rsqrt>` Returns element-wise inverse square-root value of the input
:doc:`mx.symbol.sign <./mx.symbol.sign>` Returns element-wise sign of the input
:doc:`mx.symbol.sqrt <./mx.symbol.sqrt>` Returns element-wise square-root value of the input
:doc:`mx.symbol.square <./mx.symbol.square>` Returns element-wise squared value of the input
================================================================== ==============================================================================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.ElementWiseSum
mx.symbol.abs
mx.symbol.add_n
mx.symbol.all_finite
mx.symbol.cbrt
mx.symbol.cumsum
mx.symbol.elemwise_add
mx.symbol.elemwise_div
mx.symbol.elemwise_mul
mx.symbol.elemwise_sub
mx.symbol.erf
mx.symbol.erfinv
mx.symbol.exp
mx.symbol.expm1
mx.symbol.gamma
mx.symbol.gammaln
mx.symbol.identity
mx.symbol.log
mx.symbol.log10
mx.symbol.log1p
mx.symbol.log2
mx.symbol.logical_not
mx.symbol.moments
mx.symbol.multi_all_finite
mx.symbol.multi_lars
mx.symbol.multi_sum_sq
mx.symbol.negative
mx.symbol.rcbrt
mx.symbol.reciprocal
mx.symbol.reset_arrays
mx.symbol.rsqrt
mx.symbol.sign
mx.symbol.sqrt
mx.symbol.square
Reduce
^^^^^^^^^^^^^
================================================== ===================================================================================================
:doc:`mx.symbol.max <./mx.symbol.max>` Computes the max of array elements over given axes
:doc:`mx.symbol.max_axis <./mx.symbol.max_axis>` Computes the max of array elements over given axes
:doc:`mx.symbol.mean <./mx.symbol.mean>` Computes the mean of array elements over given axes
:doc:`mx.symbol.nanprod <./mx.symbol.nanprod>` Computes the product of array elements over given axes treating Not a Numbers (``NaN``) as one
:doc:`mx.symbol.nansum <./mx.symbol.nansum>` Computes the sum of array elements over given axes treating Not a Numbers (``NaN``) as zero
:doc:`mx.symbol.prod <./mx.symbol.prod>` Computes the product of array elements over given axes
:doc:`mx.symbol.sum <./mx.symbol.sum>` Computes the sum of array elements over given axes
:doc:`mx.symbol.sum_axis <./mx.symbol.sum_axis>` Computes the sum of array elements over given axes
================================================== ===================================================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.max
mx.symbol.max_axis
mx.symbol.mean
mx.symbol.nanprod
mx.symbol.nansum
mx.symbol.prod
mx.symbol.sum
mx.symbol.sum_axis
Round
^^^^^^^^^^^^^
============================================ ==========================================================================================
:doc:`mx.symbol.ceil <./mx.symbol.ceil>` Returns element-wise ceiling of the input
:doc:`mx.symbol.clip <./mx.symbol.clip>` Clips (limits) the values in an array
:doc:`mx.symbol.fix <./mx.symbol.fix>` Returns element-wise rounded value to the nearest \ integer towards zero of the input
:doc:`mx.symbol.floor <./mx.symbol.floor>` Returns element-wise floor of the input
:doc:`mx.symbol.rint <./mx.symbol.rint>` Returns element-wise rounded value to the nearest integer of the input
:doc:`mx.symbol.round <./mx.symbol.round>` Returns element-wise rounded value to the nearest integer of the input
:doc:`mx.symbol.trunc <./mx.symbol.trunc>` Return the element-wise truncated value of the input
============================================ ==========================================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.ceil
mx.symbol.clip
mx.symbol.fix
mx.symbol.floor
mx.symbol.rint
mx.symbol.round
mx.symbol.trunc
Linear algebra
^^^^^^^^^^^^^^^^
======================================================================== ============================================================================================
:doc:`mx.symbol.L2Normalization <./mx.symbol.L2Normalization>` Normalize the input array using the L2 norm
:doc:`mx.symbol.batch_dot <./mx.symbol.batch_dot>` Batchwise dot product
:doc:`mx.symbol.dot <./mx.symbol.dot>` Dot product of two arrays
:doc:`mx.symbol.khatri_rao <./mx.symbol.khatri_rao>` Computes the Khatri-Rao product of the input matrices
:doc:`mx.symbol.linalg_det <./mx.symbol.linalg_det>` Compute the determinant of a matrix
:doc:`mx.symbol.linalg_extractdiag <./mx.symbol.linalg_extractdiag>` Extracts the diagonal entries of a square matrix
:doc:`mx.symbol.linalg_extracttrian <./mx.symbol.linalg_extracttrian>` Extracts a triangular sub-matrix from a square matrix
:doc:`mx.symbol.linalg_gelqf <./mx.symbol.linalg_gelqf>` LQ factorization for general matrix
:doc:`mx.symbol.linalg_gemm <./mx.symbol.linalg_gemm>` Performs general matrix multiplication and accumulation
:doc:`mx.symbol.linalg_gemm2 <./mx.symbol.linalg_gemm2>` Performs general matrix multiplication
:doc:`mx.symbol.linalg_inverse <./mx.symbol.linalg_inverse>` Compute the inverse of a matrix
:doc:`mx.symbol.linalg_makediag <./mx.symbol.linalg_makediag>` Constructs a square matrix with the input as diagonal
:doc:`mx.symbol.linalg_maketrian <./mx.symbol.linalg_maketrian>` Constructs a square matrix with the input representing a specific triangular sub-matrix
:doc:`mx.symbol.linalg_potrf <./mx.symbol.linalg_potrf>` Performs Cholesky factorization of a symmetric positive-definite matrix
:doc:`mx.symbol.linalg_potri <./mx.symbol.linalg_potri>` Performs matrix inversion from a Cholesky factorization
:doc:`mx.symbol.linalg_slogdet <./mx.symbol.linalg_slogdet>` Compute the sign and log of the determinant of a matrix
:doc:`mx.symbol.linalg_sumlogdiag <./mx.symbol.linalg_sumlogdiag>` Computes the sum of the logarithms of the diagonal elements of a square matrix
:doc:`mx.symbol.linalg_syrk <./mx.symbol.linalg_syrk>` Multiplication of matrix with its transpose
:doc:`mx.symbol.linalg_trmm <./mx.symbol.linalg_trmm>` Performs multiplication with a lower triangular matrix
:doc:`mx.symbol.linalg_trsm <./mx.symbol.linalg_trsm>` Solves matrix equation involving a lower triangular matrix
:doc:`mx.symbol.norm <./mx.symbol.norm>` Computes the norm on an NDArray
:doc:`mx.symbol.smooth_l1 <./mx.symbol.smooth_l1>` Calculate Smooth L1 Loss(lhs, scalar) by summing
======================================================================== ============================================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.L2Normalization
mx.symbol.batch_dot
mx.symbol.dot
mx.symbol.khatri_rao
mx.symbol.linalg_det
mx.symbol.linalg_extractdiag
mx.symbol.linalg_extracttrian
mx.symbol.linalg_gelqf
mx.symbol.linalg_gemm
mx.symbol.linalg_gemm2
mx.symbol.linalg_inverse
mx.symbol.linalg_makediag
mx.symbol.linalg_maketrian
mx.symbol.linalg_potrf
mx.symbol.linalg_potri
mx.symbol.linalg_slogdet
mx.symbol.linalg_sumlogdiag
mx.symbol.linalg_syrk
mx.symbol.linalg_trmm
mx.symbol.linalg_trsm
mx.symbol.norm
mx.symbol.smooth_l1
Trigonometric functions
^^^^^^^^^^^^^^^^^^^^^^^^^
================================================ =====================================================================
:doc:`mx.symbol.arccos <./mx.symbol.arccos>` Returns element-wise inverse cosine of the input array
:doc:`mx.symbol.arcsin <./mx.symbol.arcsin>` Returns element-wise inverse sine of the input array
:doc:`mx.symbol.arctan <./mx.symbol.arctan>` Returns element-wise inverse tangent of the input array
:doc:`mx.symbol.cos <./mx.symbol.cos>` Computes the element-wise cosine of the input array
:doc:`mx.symbol.degrees <./mx.symbol.degrees>` Converts each element of the input array from radians to degrees
:doc:`mx.symbol.radians <./mx.symbol.radians>` Converts each element of the input array from degrees to radians
:doc:`mx.symbol.sin <./mx.symbol.sin>` Computes the element-wise sine of the input array
:doc:`mx.symbol.tan <./mx.symbol.tan>` Computes the element-wise tangent of the input array
================================================ =====================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.arccos
mx.symbol.arcsin
mx.symbol.arctan
mx.symbol.cos
mx.symbol.degrees
mx.symbol.radians
mx.symbol.sin
mx.symbol.tan
Hyperbolic functions
^^^^^^^^^^^^^^^^^^^^^^^^^
================================================ ====================================================================================================
:doc:`mx.symbol.arccosh <./mx.symbol.arccosh>` Returns the element-wise inverse hyperbolic cosine of the input array, \ computed element-wise
:doc:`mx.symbol.arcsinh <./mx.symbol.arcsinh>` Returns the element-wise inverse hyperbolic sine of the input array, \ computed element-wise
:doc:`mx.symbol.arctanh <./mx.symbol.arctanh>` Returns the element-wise inverse hyperbolic tangent of the input array, \ computed element-wise
:doc:`mx.symbol.cosh <./mx.symbol.cosh>` Returns the hyperbolic cosine of the input array, computed element-wise
:doc:`mx.symbol.sinh <./mx.symbol.sinh>` Returns the hyperbolic sine of the input array, computed element-wise
:doc:`mx.symbol.tanh <./mx.symbol.tanh>` Returns the hyperbolic tangent of the input array, computed element-wise
================================================ ====================================================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.arccosh
mx.symbol.arcsinh
mx.symbol.arctanh
mx.symbol.cosh
mx.symbol.sinh
mx.symbol.tanh
Neural network symbol operations
-----------------------------------------------------
==================================================================================== ========================================================================================================
:doc:`mx.symbol.Activation <./mx.symbol.Activation>` Applies an activation function element-wise to the input
:doc:`mx.symbol.BatchNorm <./mx.symbol.BatchNorm>` Batch normalization
:doc:`mx.symbol.BatchNorm_v1 <./mx.symbol.BatchNorm_v1>` Batch normalization
:doc:`mx.symbol.BlockGrad <./mx.symbol.BlockGrad>` Stops gradient computation
:doc:`mx.symbol.CTCLoss <./mx.symbol.CTCLoss>` Connectionist Temporal Classification Loss
:doc:`mx.symbol.Convolution <./mx.symbol.Convolution>` Compute *N*-D convolution on *(N+2)*-D input
:doc:`mx.symbol.Convolution_v1 <./mx.symbol.Convolution_v1>` This operator is DEPRECATED
:doc:`mx.symbol.Correlation <./mx.symbol.Correlation>` Applies correlation to inputs
:doc:`mx.symbol.Custom <./mx.symbol.Custom>` Apply a custom operator implemented in a frontend language (like Python)
:doc:`mx.symbol.Deconvolution <./mx.symbol.Deconvolution>` Computes 1D or 2D transposed convolution (aka fractionally strided convolution) of the input tensor
:doc:`mx.symbol.Dropout <./mx.symbol.Dropout>` Applies dropout operation to input array
:doc:`mx.symbol.Embedding <./mx.symbol.Embedding>` Maps integer indices to vector representations (embeddings)
:doc:`mx.symbol.FullyConnected <./mx.symbol.FullyConnected>` Applies a linear transformation: :math:`Y = XW^T + b`
:doc:`mx.symbol.IdentityAttachKLSparseReg <./mx.symbol.IdentityAttachKLSparseReg>` Apply a sparse regularization to the output a sigmoid activation function
:doc:`mx.symbol.InstanceNorm <./mx.symbol.InstanceNorm>` Applies instance normalization to the n-dimensional input array
:doc:`mx.symbol.LRN <./mx.symbol.LRN>` Applies local response normalization to the input
:doc:`mx.symbol.LayerNorm <./mx.symbol.LayerNorm>` Layer normalization
:doc:`mx.symbol.LeakyReLU <./mx.symbol.LeakyReLU>` Applies Leaky rectified linear unit activation element-wise to the input
:doc:`mx.symbol.LinearRegressionOutput <./mx.symbol.LinearRegressionOutput>` Computes and optimizes for squared loss during backward propagation
:doc:`mx.symbol.LogisticRegressionOutput <./mx.symbol.LogisticRegressionOutput>` Applies a logistic function to the input
:doc:`mx.symbol.MAERegressionOutput <./mx.symbol.MAERegressionOutput>` Computes mean absolute error of the input
:doc:`mx.symbol.MakeLoss <./mx.symbol.MakeLoss>` Make your own loss function in network construction
:doc:`mx.symbol.Pooling <./mx.symbol.Pooling>` Performs pooling on the input
:doc:`mx.symbol.Pooling_v1 <./mx.symbol.Pooling_v1>` This operator is DEPRECATED
:doc:`mx.symbol.RNN <./mx.symbol.RNN>` Applies recurrent layers to input data
:doc:`mx.symbol.ROIPooling <./mx.symbol.ROIPooling>` Performs region of interest(ROI) pooling on the input array
:doc:`mx.symbol.SVMOutput <./mx.symbol.SVMOutput>` Computes support vector machine based transformation of the input
:doc:`mx.symbol.SoftmaxActivation <./mx.symbol.SoftmaxActivation>` Applies softmax activation to input
:doc:`mx.symbol.SoftmaxOutput <./mx.symbol.SoftmaxOutput>` Computes the gradient of cross entropy loss with respect to softmax output
:doc:`mx.symbol.SpatialTransformer <./mx.symbol.SpatialTransformer>` Applies a spatial transformer to input feature map
:doc:`mx.symbol.ctc_loss <./mx.symbol.ctc_loss>` Connectionist Temporal Classification Loss
:doc:`mx.symbol.hard_sigmoid <./mx.symbol.hard_sigmoid>` Computes hard sigmoid of x element-wise
:doc:`mx.symbol.log_softmax <./mx.symbol.log_softmax>` Computes the log softmax of the input
:doc:`mx.symbol.make_loss <./mx.symbol.make_loss>` Make your own loss function in network construction
:doc:`mx.symbol.relu <./mx.symbol.relu>` Computes rectified linear activation
:doc:`mx.symbol.sigmoid <./mx.symbol.sigmoid>` Computes sigmoid of x element-wise
:doc:`mx.symbol.softmax <./mx.symbol.softmax>` Applies the softmax function
:doc:`mx.symbol.softmax_cross_entropy <./mx.symbol.softmax_cross_entropy>` Calculate cross entropy of softmax output and one-hot label
:doc:`mx.symbol.softmin <./mx.symbol.softmin>` Applies the softmin function
:doc:`mx.symbol.softsign <./mx.symbol.softsign>` Computes softsign of x element-wise
:doc:`mx.symbol.stop_gradient <./mx.symbol.stop_gradient>` Stops gradient computation
==================================================================================== ========================================================================================================
.. toctree::
:titlesonly:
:hidden:
mx.symbol.Activation
mx.symbol.BatchNorm
mx.symbol.BatchNorm_v1
mx.symbol.BlockGrad
mx.symbol.CTCLoss
mx.symbol.Convolution
mx.symbol.Convolution_v1
mx.symbol.Correlation
mx.symbol.Custom
mx.symbol.Deconvolution
mx.symbol.Dropout
mx.symbol.Embedding
mx.symbol.FullyConnected
mx.symbol.IdentityAttachKLSparseReg
mx.symbol.InstanceNorm
mx.symbol.LRN
mx.symbol.LayerNorm
mx.symbol.LeakyReLU
mx.symbol.LinearRegressionOutput
mx.symbol.LogisticRegressionOutput
mx.symbol.MAERegressionOutput
mx.symbol.MakeLoss
mx.symbol.Pooling
mx.symbol.Pooling_v1
mx.symbol.RNN
mx.symbol.ROIPooling
mx.symbol.SVMOutput
mx.symbol.SoftmaxActivation
mx.symbol.SoftmaxOutput
mx.symbol.SpatialTransformer
mx.symbol.ctc_loss
mx.symbol.hard_sigmoid
mx.symbol.log_softmax
mx.symbol.make_loss
mx.symbol.relu
mx.symbol.sigmoid
mx.symbol.softmax
mx.symbol.softmax_cross_entropy
mx.symbol.softmin
mx.symbol.softsign
mx.symbol.stop_gradient