Expand description
Operations
Modules§
- indexing
- Indexing Arrays
Macros§
- abs
- Macro generated for the function
crate::ops::abs
. See the function documentation for more details. - acos
- Macro generated for the function
crate::ops::acos
. See the function documentation for more details. - acosh
- Macro generated for the function
crate::ops::acosh
. See the function documentation for more details. - add
- Macro generated for the function
crate::ops::add
. See the function documentation for more details. - addmm
- Macro generated for the function
crate::ops::addmm
. See the function documentation for more details. - all
- Macro generated for the function
crate::ops::all
. See the function documentation for more details. - all_
close - Macro generated for the function
crate::ops::all_close
. See the function documentation for more details. - any
- Macro generated for the function
crate::ops::any
. See the function documentation for more details. - arange
- Macro generated for the function
crate::ops::arange
. See the function documentation for more details. - argpartition
- Macro generated for the function
crate::ops::argpartition
. See the function documentation for more details. - argpartition_
all - Macro generated for the function
crate::ops::argpartition_all
. See the function documentation for more details. - argsort
- Macro generated for the function
crate::ops::argsort
. See the function documentation for more details. - argsort_
all - Macro generated for the function
crate::ops::argsort_all
. See the function documentation for more details. - array_
eq - Macro generated for the function
crate::ops::array_eq
. See the function documentation for more details. - as_
strided - Macro generated for the function
crate::ops::as_strided
. See the function documentation for more details. - asin
- Macro generated for the function
crate::ops::asin
. See the function documentation for more details. - asinh
- Macro generated for the function
crate::ops::asinh
. See the function documentation for more details. - at_
least_ 1d - Macro generated for the function
crate::ops::at_least_1d
. See the function documentation for more details. - at_
least_ 2d - Macro generated for the function
crate::ops::at_least_2d
. See the function documentation for more details. - at_
least_ 3d - Macro generated for the function
crate::ops::at_least_3d
. See the function documentation for more details. - atan
- Macro generated for the function
crate::ops::atan
. See the function documentation for more details. - atanh
- Macro generated for the function
crate::ops::atanh
. See the function documentation for more details. - block_
masked_ mm - Macro generated for the function
crate::ops::block_masked_mm
. See the function documentation for more details. - broadcast_
arrays - Macro generated for the function
crate::ops::broadcast_arrays
. See the function documentation for more details. - broadcast_
to - Macro generated for the function
crate::ops::broadcast_to
. See the function documentation for more details. - ceil
- Macro generated for the function
crate::ops::ceil
. See the function documentation for more details. - clip
- Macro generated for the function
crate::ops::clip
. See the function documentation for more details. - concatenate
- Macro generated for the function
crate::ops::concatenate
. See the function documentation for more details. - conv1d
- Macro generated for the function
crate::ops::conv1d
. See the function documentation for more details. - conv2d
- Macro generated for the function
crate::ops::conv2d
. See the function documentation for more details. - conv3d
- Macro generated for the function
crate::ops::conv3d
. See the function documentation for more details. - conv_
general - Macro generated for the function
crate::ops::conv_general
. See the function documentation for more details. - conv_
transpose1d - Macro generated for the function
crate::ops::conv_transpose1d
. See the function documentation for more details. - conv_
transpose2d - Macro generated for the function
crate::ops::conv_transpose2d
. See the function documentation for more details. - conv_
transpose3d - Macro generated for the function
crate::ops::conv_transpose3d
. See the function documentation for more details. - cos
- Macro generated for the function
crate::ops::cos
. See the function documentation for more details. - cosh
- Macro generated for the function
crate::ops::cosh
. See the function documentation for more details. - cummax
- Macro generated for the function
crate::ops::cummax
. See the function documentation for more details. - cummin
- Macro generated for the function
crate::ops::cummin
. See the function documentation for more details. - cumprod
- Macro generated for the function
crate::ops::cumprod
. See the function documentation for more details. - cumsum
- Macro generated for the function
crate::ops::cumsum
. See the function documentation for more details. - degrees
- Macro generated for the function
crate::ops::degrees
. See the function documentation for more details. - dequantize
- Macro generated for the function
crate::ops::dequantize
. See the function documentation for more details. - diag
- Macro generated for the function
crate::ops::diag
. See the function documentation for more details. - diagonal
- Macro generated for the function
crate::ops::diagonal
. See the function documentation for more details. - divide
- Macro generated for the function
crate::ops::divide
. See the function documentation for more details. - divmod
- Macro generated for the function
crate::ops::divmod
. See the function documentation for more details. - einsum
- Macro generated for the function
crate::ops::einsum
. See the function documentation for more details. - eq
- Macro generated for the function
crate::ops::eq
. See the function documentation for more details. - erf
- Macro generated for the function
crate::ops::erf
. See the function documentation for more details. - erfinv
- Macro generated for the function
crate::ops::erfinv
. See the function documentation for more details. - exp
- Macro generated for the function
crate::ops::exp
. See the function documentation for more details. - expand_
dims - Macro generated for the function
crate::ops::expand_dims
. See the function documentation for more details. - expm1
- Macro generated for the function
crate::ops::expm1
. See the function documentation for more details. - eye
- Macro generated for the function
crate::ops::eye
. See the function documentation for more details. - flatten
- Macro generated for the function
crate::ops::flatten
. See the function documentation for more details. - floor
- Macro generated for the function
crate::ops::floor
. See the function documentation for more details. - floor_
divide - Macro generated for the function
crate::ops::floor_divide
. See the function documentation for more details. - full
- Macro generated for the function
crate::ops::full
. See the function documentation for more details. - ge
- Macro generated for the function
crate::ops::ge
. See the function documentation for more details. - gt
- Macro generated for the function
crate::ops::gt
. See the function documentation for more details. - identity
- Macro generated for the function
crate::ops::identity
. See the function documentation for more details. - inner
- Macro generated for the function
crate::ops::inner
. See the function documentation for more details. - is_
close - Macro generated for the function
crate::ops::is_close
. See the function documentation for more details. - is_inf
- Macro generated for the function
crate::ops::is_inf
. See the function documentation for more details. - is_nan
- Macro generated for the function
crate::ops::is_nan
. See the function documentation for more details. - is_
neg_ inf - Macro generated for the function
crate::ops::is_neg_inf
. See the function documentation for more details. - is_
pos_ inf - Macro generated for the function
crate::ops::is_pos_inf
. See the function documentation for more details. - kron
- Macro generated for the function
crate::ops::kron
. See the function documentation for more details. - le
- Macro generated for the function
crate::ops::le
. See the function documentation for more details. - linspace
- Macro generated for the function
crate::ops::linspace
. See the function documentation for more details. - log
- Macro generated for the function
crate::ops::log
. See the function documentation for more details. - log2
- Macro generated for the function
crate::ops::log2
. See the function documentation for more details. - log1p
- Macro generated for the function
crate::ops::log1p
. See the function documentation for more details. - log10
- Macro generated for the function
crate::ops::log10
. See the function documentation for more details. - log_
add_ exp - Macro generated for the function
crate::ops::log_add_exp
. See the function documentation for more details. - log_
sum_ exp - Macro generated for the function
crate::ops::log_sum_exp
. See the function documentation for more details. - logical_
and - Macro generated for the function
crate::ops::logical_and
. See the function documentation for more details. - logical_
not - Macro generated for the function
crate::ops::logical_not
. See the function documentation for more details. - logical_
or - Macro generated for the function
crate::ops::logical_or
. See the function documentation for more details. - lt
- Macro generated for the function
crate::ops::lt
. See the function documentation for more details. - matmul
- Macro generated for the function
crate::ops::matmul
. See the function documentation for more details. - max
- Macro generated for the function
crate::ops::max
. See the function documentation for more details. - maximum
- Macro generated for the function
crate::ops::maximum
. See the function documentation for more details. - mean
- Macro generated for the function
crate::ops::mean
. See the function documentation for more details. - min
- Macro generated for the function
crate::ops::min
. See the function documentation for more details. - minimum
- Macro generated for the function
crate::ops::minimum
. See the function documentation for more details. - move_
axis - Macro generated for the function
crate::ops::move_axis
. See the function documentation for more details. - multiply
- Macro generated for the function
crate::ops::multiply
. See the function documentation for more details. - ne
- Macro generated for the function
crate::ops::ne
. See the function documentation for more details. - negative
- Macro generated for the function
crate::ops::negative
. See the function documentation for more details. - ones
- Macro generated for the function
crate::ops::ones
. See the function documentation for more details. - ones_
dtype - Macro generated for the function
crate::ops::ones_dtype
. See the function documentation for more details. - ones_
like - Macro generated for the function
crate::ops::ones_like
. See the function documentation for more details. - outer
- Macro generated for the function
crate::ops::outer
. See the function documentation for more details. - pad
- Macro generated for the function
crate::ops::pad
. See the function documentation for more details. - partition
- Macro generated for the function
crate::ops::partition
. See the function documentation for more details. - partition_
all - Macro generated for the function
crate::ops::partition_all
. See the function documentation for more details. - power
- Macro generated for the function
crate::ops::power
. See the function documentation for more details. - prod
- Macro generated for the function
crate::ops::prod
. See the function documentation for more details. - quantize
- Macro generated for the function
crate::ops::quantize
. See the function documentation for more details. - quantized_
matmul - Macro generated for the function
crate::ops::quantized_matmul
. See the function documentation for more details. - radians
- Macro generated for the function
crate::ops::radians
. See the function documentation for more details. - reciprocal
- Macro generated for the function
crate::ops::reciprocal
. See the function documentation for more details. - remainder
- Macro generated for the function
crate::ops::remainder
. See the function documentation for more details. - repeat
- Macro generated for the function
crate::ops::repeat
. See the function documentation for more details. - repeat_
all - Macro generated for the function
crate::ops::repeat_all
. See the function documentation for more details. - reshape
- Macro generated for the function
crate::ops::reshape
. See the function documentation for more details. - round
- Macro generated for the function
crate::ops::round
. See the function documentation for more details. - rsqrt
- Macro generated for the function
crate::ops::rsqrt
. See the function documentation for more details. - sigmoid
- Macro generated for the function
crate::ops::sigmoid
. See the function documentation for more details. - sign
- Macro generated for the function
crate::ops::sign
. See the function documentation for more details. - sin
- Macro generated for the function
crate::ops::sin
. See the function documentation for more details. - sinh
- Macro generated for the function
crate::ops::sinh
. See the function documentation for more details. - softmax
- Macro generated for the function
crate::ops::softmax
. See the function documentation for more details. - softmax_
all - Macro generated for the function
crate::ops::softmax_all
. See the function documentation for more details. - sort
- Macro generated for the function
crate::ops::sort
. See the function documentation for more details. - sort_
all - Macro generated for the function
crate::ops::sort_all
. See the function documentation for more details. - split
- Macro generated for the function
crate::ops::split
. See the function documentation for more details. - split_
equal - Macro generated for the function
crate::ops::split_equal
. See the function documentation for more details. - sqrt
- Macro generated for the function
crate::ops::sqrt
. See the function documentation for more details. - square
- Macro generated for the function
crate::ops::square
. See the function documentation for more details. - squeeze
- Macro generated for the function
crate::ops::squeeze
. See the function documentation for more details. - stack
- Macro generated for the function
crate::ops::stack
. See the function documentation for more details. - stack_
all - Macro generated for the function
crate::ops::stack_all
. See the function documentation for more details. - std
- Macro generated for the function
crate::ops::std
. See the function documentation for more details. - subtract
- Macro generated for the function
crate::ops::subtract
. See the function documentation for more details. - sum
- Macro generated for the function
crate::ops::sum
. See the function documentation for more details. - swap_
axes - Macro generated for the function
crate::ops::swap_axes
. See the function documentation for more details. - tan
- Macro generated for the function
crate::ops::tan
. See the function documentation for more details. - tanh
- Macro generated for the function
crate::ops::tanh
. See the function documentation for more details. - tensordot
- Macro generated for the function
crate::ops::tensordot
. See the function documentation for more details. - tile
- Macro generated for the function
crate::ops::tile
. See the function documentation for more details. - transpose
- Macro generated for the function
crate::ops::transpose
. See the function documentation for more details. - transpose_
all - Macro generated for the function
crate::ops::transpose_all
. See the function documentation for more details. - tri
- Macro generated for the function
crate::ops::tri
. See the function documentation for more details. - tril
- Macro generated for the function
crate::ops::tril
. See the function documentation for more details. - triu
- Macro generated for the function
crate::ops::triu
. See the function documentation for more details. - unflatten
- Macro generated for the function
crate::ops::unflatten
. See the function documentation for more details. - variance
- Macro generated for the function
crate::ops::variance
. See the function documentation for more details. - which
- Macro generated for the function
crate::ops::which
. See the function documentation for more details. - zeros
- Macro generated for the function
crate::ops::zeros
. See the function documentation for more details. - zeros_
dtype - Macro generated for the function
crate::ops::zeros_dtype
. See the function documentation for more details. - zeros_
like - Macro generated for the function
crate::ops::zeros_like
. See the function documentation for more details.
Enums§
- PadMode
- The padding mode.
- PadWidth
- Number of padding values to add to the edges of each axis.
- Tensor
DotDims - The number of dimensions to sum over
Traits§
- Clip
Bound - A custom trait for the bound of the clip operation.
Functions§
- abs
- Element-wise absolute value.
- abs_
device - Element-wise absolute value.
- acos
- Element-wise inverse cosine.
- acos_
device - Element-wise inverse cosine.
- acosh
- Element-wise inverse hyperbolic cosine.
- acosh_
device - Element-wise inverse hyperbolic cosine.
- add
- See
Array::add
. - add_
device - See
Array::add
. - addmm
- Matrix multiplication with addition and optional scaling.
- addmm_
device - Matrix multiplication with addition and optional scaling.
- all
- See
Array::all
- all_
close - See
Array::all_close
- all_
close_ device - See
Array::all_close
- all_
device - See
Array::all
- any
- See
Array::any
- any_
device - See
Array::any
- arange
- See
Array::arange
- arange_
device - See
Array::arange
- argpartition
- Returns the indices that partition the array. Returns an error if the arguments are invalid.
- argpartition_
all - Returns the indices that partition the flattened array. Returns an error if the arguments are invalid.
- argpartition_
all_ device - Returns the indices that partition the flattened array. Returns an error if the arguments are invalid.
- argpartition_
device - Returns the indices that partition the array. Returns an error if the arguments are invalid.
- argsort
- Returns the indices that sort the array. Returns an error if the arguments are invalid.
- argsort_
all - Returns the indices that sort the flattened array. Returns an error if the arguments are invalid.
- argsort_
all_ device - Returns the indices that sort the flattened array. Returns an error if the arguments are invalid.
- argsort_
device - Returns the indices that sort the array. Returns an error if the arguments are invalid.
- array_
eq - See
Array::array_eq
- array_
eq_ device - See
Array::array_eq
- as_
strided - Create a view into the array with the given shape and strides.
- as_
strided_ device - Create a view into the array with the given shape and strides.
- asin
- Element-wise inverse sine.
- asin_
device - Element-wise inverse sine.
- asinh
- Element-wise inverse hyperbolic sine.
- asinh_
device - Element-wise inverse hyperbolic sine.
- at_
least_ 1d - Convert array to have at least one dimension.
- at_
least_ 1d_ device - Convert array to have at least one dimension.
- at_
least_ 2d - Convert array to have at least two dimensions.
- at_
least_ 2d_ device - Convert array to have at least two dimensions.
- at_
least_ 3d - Convert array to have at least three dimensions.
- at_
least_ 3d_ device - Convert array to have at least three dimensions.
- atan
- Element-wise inverse tangent.
- atan_
device - Element-wise inverse tangent.
- atanh
- Element-wise inverse hyperbolic tangent.
- atanh_
device - Element-wise inverse hyperbolic tangent.
- block_
masked_ mm - Matrix multiplication with block masking.
- block_
masked_ mm_ device - Matrix multiplication with block masking.
- broadcast_
arrays - Broadcast a vector of arrays against one another. Returns an error if the shapes are broadcastable.
- broadcast_
arrays_ device - Broadcast a vector of arrays against one another. Returns an error if the shapes are broadcastable.
- broadcast_
to - Broadcast an array to the given shape. Returns an error if the shapes are not broadcastable.
- broadcast_
to_ device - Broadcast an array to the given shape. Returns an error if the shapes are not broadcastable.
- ceil
- Element-wise ceiling.
- ceil_
device - Element-wise ceiling.
- clip
- Clip the values of the array between the given minimum and maximum.
- clip_
device - Clip the values of the array between the given minimum and maximum.
- concatenate
- Concatenate the arrays along the given axis. Returns an error if the shapes are invalid.
- concatenate_
device - Concatenate the arrays along the given axis. Returns an error if the shapes are invalid.
- conv1d
- 1D convolution over an input with several channels returning an error if the inputs are invalid.
- conv1d_
device - 1D convolution over an input with several channels returning an error if the inputs are invalid.
- conv2d
- 2D convolution over an input with several channels returning an error if the inputs are invalid.
- conv2d_
device - 2D convolution over an input with several channels returning an error if the inputs are invalid.
- conv3d
- 3D convolution over an input with several channels.
- conv3d_
device - 3D convolution over an input with several channels.
- conv_
general - General convolution over an input with several channels returning an error if the inputs are invalid.
- conv_
general_ device - General convolution over an input with several channels returning an error if the inputs are invalid.
- conv_
transpose1d - 1D transposed convolution over an input with several channels.
- conv_
transpose1d_ device - 1D transposed convolution over an input with several channels.
- conv_
transpose2d - 2D transposed convolution over an input with several channels.
- conv_
transpose2d_ device - 2D transposed convolution over an input with several channels.
- conv_
transpose3d - 3D transposed convolution over an input with several channels.
- conv_
transpose3d_ device - 3D transposed convolution over an input with several channels.
- cos
- Element-wise cosine.
- cos_
device - Element-wise cosine.
- cosh
- Element-wise hyperbolic cosine.
- cosh_
device - Element-wise hyperbolic cosine.
- cummax
- See
Array::cummax
- cummax_
device - See
Array::cummax
- cummin
- See
Array::cummin
- cummin_
device - See
Array::cummin
- cumprod
- See
Array::cumprod
- cumprod_
device - See
Array::cumprod
- cumsum
- See
Array::cumsum
- cumsum_
device - See
Array::cumsum
- degrees
- Convert angles from radians to degrees.
- degrees_
device - Convert angles from radians to degrees.
- dequantize
- Dequantize the matrix
w
using the providedscales
andbiases
and thegroup_size
andbits
configuration. - dequantize_
device - Dequantize the matrix
w
using the providedscales
andbiases
and thegroup_size
andbits
configuration. - diag
- See
Array::diag
- diag_
device - See
Array::diag
- diagonal
- See
Array::diagonal
- diagonal_
device - See
Array::diagonal
- divide
- See
Array::divide
. - divide_
device - See
Array::divide
. - divmod
- Element-wise quotient and remainder.
- divmod_
device - Element-wise quotient and remainder.
- einsum
- Perform the Einstein summation convention on the operands.
- einsum_
device - Perform the Einstein summation convention on the operands.
- eq
- See
Array::eq
- eq_
device - See
Array::eq
- erf
- Element-wise error function.
- erf_
device - Element-wise error function.
- erfinv
- Element-wise inverse error function.
- erfinv_
device - Element-wise inverse error function.
- exp
- See
Array::exp
. - exp_
device - See
Array::exp
. - expand_
dims - Add a size one dimension at the given axis, returns an error if the axes are invalid.
- expand_
dims_ device - Add a size one dimension at the given axis, returns an error if the axes are invalid.
- expm1
- Element-wise exponential minus 1.
- expm1_
device - Element-wise exponential minus 1.
- eye
- See
Array::eye
- eye_
device - See
Array::eye
- flatten
- Flatten an array. Returns an error if the axes are invalid.
- flatten_
device - Flatten an array. Returns an error if the axes are invalid.
- floor
- See
Array::floor
. - floor_
device - See
Array::floor
. - floor_
divide - See
Array::floor_divide
. - floor_
divide_ device - See
Array::floor_divide
. - full
- See
Array::full
- full_
device - See
Array::full
- ge
- See
Array::ge
- ge_
device - See
Array::ge
- gt
- See
Array::gt
- gt_
device - See
Array::gt
- identity
- See
Array::identity
- identity_
device - See
Array::identity
- inner
- Ordinary inner product of vectors for 1-D arrays, in higher dimensions a sum product over the last axes.
- inner_
device - Ordinary inner product of vectors for 1-D arrays, in higher dimensions a sum product over the last axes.
- is_
close - See
Array::is_close
- is_
close_ device - See
Array::is_close
- is_inf
- Return a boolean array indicating which elements are +/- inifnity.
- is_
inf_ device - Return a boolean array indicating which elements are +/- inifnity.
- is_nan
- Return a boolean array indicating which elements are NaN.
- is_
nan_ device - Return a boolean array indicating which elements are NaN.
- is_
neg_ inf - Return a boolean array indicating which elements are negative infinity.
- is_
neg_ inf_ device - Return a boolean array indicating which elements are negative infinity.
- is_
pos_ inf - Return a boolean array indicating which elements are positive infinity.
- is_
pos_ inf_ device - Return a boolean array indicating which elements are positive infinity.
- kron
- Perform the Kronecker product of two arrays.
- kron_
device - Perform the Kronecker product of two arrays.
- le
- See
Array::le
- le_
device - See
Array::le
- linspace
- See
Array::linspace
- linspace_
device - See
Array::linspace
- log
- See
Array::log
. - log2
- See
Array::log2
. - log1p
- See
Array::log1p
. - log1p_
device - See
Array::log1p
. - log2_
device - See
Array::log2
. - log10
- See
Array::log10
. - log10_
device - See
Array::log10
. - log_
add_ exp - Element-wise log-add-exp.
- log_
add_ exp_ device - Element-wise log-add-exp.
- log_
device - See
Array::log
. - log_
sum_ exp - See
Array::log_sum_exp
- log_
sum_ exp_ device - See
Array::log_sum_exp
- logical_
and - See
Array::logical_and
- logical_
and_ device - See
Array::logical_and
- logical_
not - See
Array::logical_not
- logical_
not_ device - See
Array::logical_not
- logical_
or - See
Array::logical_or
- logical_
or_ device - See
Array::logical_or
- lt
- See
Array::lt
- lt_
device - See
Array::lt
- matmul
- See
Array::matmul
. - matmul_
device - See
Array::matmul
. - max
- See
Array::max
- max_
device - See
Array::max
- maximum
- Element-wise maximum.
- maximum_
device - Element-wise maximum.
- mean
- See
Array::mean
- mean_
device - See
Array::mean
- min
- See
Array::min
- min_
device - See
Array::min
- minimum
- Element-wise minimum.
- minimum_
device - Element-wise minimum.
- move_
axis - Move an axis to a new position. Returns an error if the axes are invalid.
- move_
axis_ device - Move an axis to a new position. Returns an error if the axes are invalid.
- multiply
- See
Array::multiply
. - multiply_
device - See
Array::multiply
. - ne
- See
Array::ne
- ne_
device - See
Array::ne
- negative
- See
Array::negative
. - negative_
device - See
Array::negative
. - ones
- See
Array::ones
- ones_
device - See
Array::ones
- ones_
dtype - Similar to
Array::ones
but with a specified dtype. - ones_
dtype_ device - Similar to
Array::ones
but with a specified dtype. - ones_
like - An array of ones like the input.
- ones_
like_ device - An array of ones like the input.
- outer
- Compute the outer product of two 1-D arrays, if the array’s passed are not 1-D a flatten op will be run beforehand.
- outer_
device - Compute the outer product of two 1-D arrays, if the array’s passed are not 1-D a flatten op will be run beforehand.
- pad
- Pad an array with a constant value. Returns an error if the width is invalid.
- pad_
device - Pad an array with a constant value. Returns an error if the width is invalid.
- partition
- Returns a partitioned copy of the array such that the smaller
kth
elements are first. Returns an error if the arguments are invalid. - partition_
all - Returns a partitioned copy of the flattened array such that the smaller
kth
elements are first. Returns an error if the arguments are invalid. - partition_
all_ device - Returns a partitioned copy of the flattened array such that the smaller
kth
elements are first. Returns an error if the arguments are invalid. - partition_
device - Returns a partitioned copy of the array such that the smaller
kth
elements are first. Returns an error if the arguments are invalid. - power
- See
Array::power
. - power_
device - See
Array::power
. - prod
- See
Array::prod
- prod_
device - See
Array::prod
- quantize
- Quantize the matrix
w
usingbits
bits per element. - quantize_
device - Quantize the matrix
w
usingbits
bits per element. - quantized_
matmul - Perform the matrix multiplication with the quantized matrix
w
. The quantization uses one floating point scale and bias pergroup_size
of elements. Each element inw
takesbits
bits and is packed in an unsigned 32 bit integer. - quantized_
matmul_ device - Perform the matrix multiplication with the quantized matrix
w
. The quantization uses one floating point scale and bias pergroup_size
of elements. Each element inw
takesbits
bits and is packed in an unsigned 32 bit integer. - radians
- Convert angles from degrees to radians.
- radians_
device - Convert angles from degrees to radians.
- reciprocal
- See
Array::reciprocal
. - reciprocal_
device - See
Array::reciprocal
. - remainder
- See
Array::remainder
. - remainder_
device - See
Array::remainder
. - repeat
- See
Array::repeat
- repeat_
all - See
Array::repeat_all
- repeat_
all_ device - See
Array::repeat_all
- repeat_
device - See
Array::repeat
- reshape
- Reshape an array while preserving the size. Returns an error if the new shape is invalid.
- reshape_
device - Reshape an array while preserving the size. Returns an error if the new shape is invalid.
- round
- See
Array::round
. - round_
device - See
Array::round
. - rsqrt
- See
Array::rsqrt
. - rsqrt_
device - See
Array::rsqrt
. - sigmoid
- Element-wise logistic sigmoid.
- sigmoid_
device - Element-wise logistic sigmoid.
- sign
- Element-wise sign.
- sign_
device - Element-wise sign.
- sin
- See
Array::sin
. - sin_
device - See
Array::sin
. - sinh
- Element-wise hyperbolic sine.
- sinh_
device - Element-wise hyperbolic sine.
- softmax
- Perform the softmax along the given axis.
- softmax_
all - Perform the softmax along all axes.
- softmax_
all_ device - Perform the softmax along all axes.
- softmax_
device - Perform the softmax along the given axis.
- sort
- Returns a sorted copy of the array. Returns an error if the arguments are invalid.
- sort_
all - Returns a sorted copy of the flattened array. Returns an error if the arguments are invalid.
- sort_
all_ device - Returns a sorted copy of the flattened array. Returns an error if the arguments are invalid.
- sort_
device - Returns a sorted copy of the array. Returns an error if the arguments are invalid.
- split
- Split an array along a given axis. Returns an error if the indices are invalid.
- split_
device - Split an array along a given axis. Returns an error if the indices are invalid.
- split_
equal - Split an array into equal parts along a given axis. Returns an error if the array cannot be split into equal parts.
- split_
equal_ device - Split an array into equal parts along a given axis. Returns an error if the array cannot be split into equal parts.
- sqrt
- See
Array::sqrt
. - sqrt_
device - See
Array::sqrt
. - square
- See
Array::square
. - square_
device - See
Array::square
. - squeeze
- Remove length one axes from an array. Returns an error if the axes are invalid.
- squeeze_
device - Remove length one axes from an array. Returns an error if the axes are invalid.
- stack
- Stacks the arrays along a new axis. Returns an error if the arguments are invalid.
- stack_
all - Stacks the arrays along a new axis. Returns an error if the arrays have different shapes.
- stack_
all_ device - Stacks the arrays along a new axis. Returns an error if the arrays have different shapes.
- stack_
device - Stacks the arrays along a new axis. Returns an error if the arguments are invalid.
- std
- Compute the standard deviation(s) over the given axes.
- std_
device - Compute the standard deviation(s) over the given axes.
- subtract
- See
Array::subtract
. - subtract_
device - See
Array::subtract
. - sum
- See
Array::sum
- sum_
device - See
Array::sum
- swap_
axes - Swap two axes of an array. Returns an error if the axes are invalid.
- swap_
axes_ device - Swap two axes of an array. Returns an error if the axes are invalid.
- tan
- See [
Array::tan
]. - tan_
device - See [
Array::tan
]. - tanh
- Element-wise hyperbolic tangent.
- tanh_
device - Element-wise hyperbolic tangent.
- tensordot
- Compute the tensor dot product along the specified axes.
- tensordot_
device - Compute the tensor dot product along the specified axes.
- tile
- Construct an array by repeating
a
the number of times given byreps
. - tile_
device - Construct an array by repeating
a
the number of times given byreps
. - transpose
- Transpose the dimensions of the array. Returns an error if the axes are invalid.
- transpose_
all - Transpose with all axes reversed
- transpose_
all_ device - Transpose with all axes reversed
- transpose_
device - Transpose the dimensions of the array. Returns an error if the axes are invalid.
- tri
- See
Array::tri
- tri_
device - See
Array::tri
- tril
- Zeros the array above the given diagonal
- tril_
device - Zeros the array above the given diagonal
- triu
- Zeros the array below the given diagonal
- triu_
device - Zeros the array below the given diagonal
- unflatten
- Unflatten an axis of an array to a shape.
- unflatten_
device - Unflatten an axis of an array to a shape.
- variance
- See
Array::variance
- variance_
device - See
Array::variance
- where
- Select from
a
orb
according tocondition
returning an error if the arrays are not broadcastable. - where_
device - Select from
a
orb
according tocondition
returning an error if the arrays are not broadcastable. - which
- Alias for [
r#where
] - which_
device - Alias for [
r#where
] - zeros
- See
Array::zeros
- zeros_
device - See
Array::zeros
- zeros_
dtype - Similar to
Array::zeros
but with a specified dtype. - zeros_
dtype_ device - Similar to
Array::zeros
but with a specified dtype. - zeros_
like - An array of zeros like the input.
- zeros_
like_ device - An array of zeros like the input.