Crate arrayfire

Expand description

ArrayFire is a high performance software library for parallel computing with an easy-to-use API. ArrayFire abstracts away much of the details of programming parallel architectures by providing a high-level container object, the Array, that represents data stored on a CPU, GPU, FPGA, or other type of accelerator. This abstraction permits developers to write massively parallel applications in a high-level language where they need not be concerned about low-level optimizations that are frequently required to achieve high throughput on most parallel architectures. This crate provides Rust bindings for the ArrayFire library. Given below table shows the rust bindings compatability with ArrayFire upstream. If you find any bugs, please report them on github.

arrayfire-rust crate	ArrayFire Upstream
M.m.p1	M.m.p2

Only, Major(M) & Minor(m) version numbers need to match. p1 and p2 are patch/fix updates for arrayfire-rust & ArrayFire respectively, and they don’t need to match.

Please go through our tutorials book for more explanations on how to use ArrayFire to speedup your code.

Macros§

af_print: Print given message before printing out the Array to standard output
constant: Create an array of given shape filled with a single value a.k.a constant array
dim4: Create a dim4 object from provided dimensions
eval: Macro to evaluate individual Arrays or assignment operations
join_many: Join multiple Arrays along a given dimension
mem_info: Macro to print the current stats of ArrayFire’s memory manager.
randn: Create an array of given shape sampled from normal distribution
randu: Create an array of given shape sampled from uniform distribution
seq: Create a sequence object
view: Indexing into an existing Array

Structs§

Array: A multidimensional data container
Callback: Structure holding handle to callback function
Dim4: Dim4 is used to store Array dimensions
Event: RAII construct to manage ArrayFire events
Features: A set of Array objects (usually, used in Computer vision context)
Indexer: Struct to manage an array of resources of type af_indexer_t(ArrayFire C struct)
RandomEngine: Random number generator engine
Seq: Sequences are used for indexing Arrays
Window: Used to render Array objects

Enums§

AfError: Error codes
Backend: Compute/Acceleration Backend
BinaryOp: Binary operation types for generalized scan functions
BorderType: Helps determine how to pad kernels along borders
CannyThresholdType: Canny edge detector threshold operations types
ColorMap: Dictates what color map is used for Image rendering
ColorSpace: Identify the color space of given image(Array)
Connectivity: Used by regions function to identify type of connectivity
ConvDomain: Helps determine if convolution is in Spatial or Frequency domain
ConvGradientType: Gradient mode for convolution
ConvMode: Helps determine the size of output of convolution
CublasMathMode: Gradient mode for convolution
DType: Types of Array data type
DiffusionEq: Anisotropic diffusion flux equation types
FluxFn: Diffusion equation types
HomographyType: Homography type
InterpType: Dictates the interpolation method to be used by a function
InverseDeconvAlgo: Inverse Deconvolution Algorithm
IterativeDeconvAlgo: Iterative Deconvolution Algorithm
MarkerType: Plotting markers
MatProp: Helps determine the type of a Matrix
MatchType: Error metric used by matchTemplate function
MomentType: Image moment types
NormType: Norm type
RandomEngineType: Random engine types
Scalar: Scalar value types
SparseFormat: Sparse storage format type
TopkFn: topk function ordering
VarianceBias: Gradient mode for convolution
YCCStd: YCbCr Standards

Constants§

DEFAULT_RANDOM_ENGINE: Default RandomEngine that defaults to PHILOX
MERSENNE: Default Mersenne RandomEngine that points to MERSENNE_GP11213
PHILOX: Default Philon RandomEngine that points to PHILOX_4X32_10
THREEFRY: Default Threefry RandomEngine that points to THREEFRY_2X32_16

Traits§

ComplexFloating: Trait qualifier to accept complex data(numbers)
ConfidenceCCInput: Trait qualifier for confidence connected components input
ConstGenerator: Type Trait to generate a constant Array of given size
Convertable: Type Trait to convert to an Array
CovarianceComputable: Trait qualifier for given type indicating computability of covariance
DeconvInput: Trait qualifier for confidence connected components input
EdgeComputable: Trait qualifier for given type indicating if edge calculations such as derivates etc. can be performed
FloatingPoint: Trait qualifier to accept either real or complex typed data
Fromf64: Trait to convert reduction’s scalar output to appropriate output type
GrayRGBConvertible: Trait qualifier for given type indicating conversion capability between grayscale and RGB triplets of data
HasAfEnum: Types of the data that can be generated using ArrayFire data generation functions.
ImageFilterType: Trait qualifier for type of Array’s that are accepted by image processing functions especially filtering algorithms
ImageNativeType: Trait qualifier for type of Array’s that are accepted by native image load/save functions.
ImplicitPromote: This is an internal trait defined and implemented by ArrayFire create for rust’s built-in types to figure out the data type binary operation’s results.
Indexable: Trait bound indicating indexability
IndexableType: Trait qualifier for the type of Arrays accepted by scan operations
IntegralType: Trait qualifier for given type indicating computability of covariance
MedianComputable: Trait qualifier for given type indicating computability of Median
MomentsComputable: Trait qualifier for given type indicating computability of Moments
RealFloating: Trait qualifier to accept real data(numbers)
RealNumber: Trait qualifier indicating it can hold real numbers only
ReduceByKeyInput: Trait qualifier for Reduction Key type
Scanable: Trait qualifier for the type of Arrays accepted by scan operations

Functions§

HANDLE_ERROR: Default error handler for error code returned by ArrayFire FFI calls
abs: Computes absolute value
accum: Perform exclusive sum of elements along a given dimension
acos: Compute acos
acosh: Compute acosh
add: Addition of two Arrays
all_true: Find if all of the values along a given dimension in the Array are true
all_true_all: Find if all values of Array are non-zero
all_true_by_key: Key based AND of elements along a given dimension
alloc_host: Allocates space using Arrayfire allocator in host memory
alloc_pinned^⚠: Allocate non-pageable memory on HOST memory
and: Elementwise logical and operation of two Arrays
anisotropic_diffusion: Anisotropic smoothing filter
any_true: Find if any of the values along a given dimension in the Array are true
any_true_all: Find if any value of Array is non-zero
any_true_by_key: Key based OR of elements along a given dimension
approx1: Perform signal interpolation for 1d signals
approx2: Perform signal interpolation for 2d signals
approx1_uniform: Perform signal interpolation for 1d signals along specified dimension
approx1_uniform_v2: Same as approx1_uniform but uses existing Array as output
approx1_v2: Same as approx1 but uses existing Array as output
approx2_uniform: Perform signal interpolation for 2d signals along a specified dimension
approx2_uniform_v2: Same as approx2_uniform but uses existing Array as output
approx2_v2: Same as approx2 but uses existing Array as output
arg: Computes phase value
asin: Compute asin
asinh: Compute asinh
assign_gen: Assign an Array to another after indexing it using any combination of Array’s and Sequence’s
assign_seq: Assign(copy) content of an Array to another Array indexed by Sequences
atan: Compute atan
atan2: Calculate atan2 of two Arrays
atanh: Compute atanh
bilateral: Bilateral Filter.
bitand: Elementwise AND(bit) operation of two Arrays
bitnot: Perform bitwise complement on all values of Array
bitor: Elementwise OR(bit) operation of two Arrays
bitxor: Elementwise XOR(bit) operation of two Arrays
canny: Canny edge detection operator
cbrt: Compute the cube root
ceil: Ceil the values in an Array
cholesky: Perform Cholesky decomposition
cholesky_inplace: Perform inplace Cholesky decomposition
clamp: Clamp the values of Array
col: Extract col_num col from input Array
color_space: Color space conversion
cols: Get all cols from first to last in the input Array
confidence_cc: Segment image based on similar pixel characteristics
conjg: Compute the complex conjugate
constant: Create an Array with constant value
convolve1: 1d convolution
convolve2: 2d convolution
convolve3: 3d convolution
convolve2_gradient_nn: Backward pass gradient of 2D convolution
convolve2_nn: Convolution Integral for two dimensional data
convolve2_sep: Separable convolution for 2d signals
corrcoef: Compute correlation coefficient
cos: Compute cos
cosh: Compute cosh
count: Count number of non-zero elements along a given dimension
count_all: Count number of non-zero values in the Array
count_by_key: Find total count of elements with similar keys along a given dimension
covDeprecated: Compute covariance of two Arrays
cov_v2: Compute covariance of two Arrays
cplx: Create a complex Array from real Array
cplx2: Create complex array from two Arrays
det: Find the determinant of the matrix
device_count: Get total number of available devices
device_gc: Call the garbage collection routine
device_info: Gets the information about device and platform as strings.
device_mem_info: Get memory information from the memory manager for the current active device
diag_create: Create a diagonal matrix
diag_extract: Extract diagonal from a given Matrix
diff1: Calculate first order numerical difference along a given dimension
diff2: Calculate second order numerical difference along a given dimension
dilate: Dilate an Image
dilate3: Dilate a Volume
div: Division of two Arrays
dog: Difference of Gaussians.
dot: Calculate the dot product of vectors.
eq: Perform equals comparison operation
erf: Compute error function value
erfc: Compute the complementary error function value
erode: Erode an Image
erode3: Erode a Volume
eval_multiple: evaluate multiple arrays
exp: Compute e raised to the power of value
expm1: Compute e raised to the power of value -1
factorial: Compute the factorial
fast: Fast feature detector
fft: Fast fourier transform for 1d signals
fft2: Fast fourier transform for 2d signals
fft3: Fast fourier transform for 3d signals
fft2_c2r: 2d Complex to Real fast fourier transform
fft2_inplace: In place 2d dimensional Fast fourier transform
fft2_r2c: 2d Real to Complex fast fourier transform
fft3_c2r: 3d Complex to Real fast fourier transform
fft3_inplace: In place 3d dimensional Fast fourier transform
fft3_r2c: 3d Real to Complex fast fourier transform
fft_c2r: 1d Complex to Real fast fourier transform
fft_convolve1: 1d convolution using fast-fourier transform
fft_convolve2: 2d convolution using fast-fourier transform
fft_convolve3: 3d convolution using fast-fourier transform
fft_inplace: In place 1d dimensional Fast fourier transform
fft_r2c: 1d Real to Complex fast fourier transform
fir: Finite impulse filter
flat: Flatten the multidimensional Array to an 1D Array
flip: Flip the Array
floor: Floor the values in an Array
free_host: Frees memory allocated by Arrayfire allocator in host memory
free_pinned^⚠: Free the pointer returned by alloc_pinned
gaussian_kernel: Creates a Gaussian Kernel.
ge: Perform greater than equals comparison operation
gemm: BLAS general matrix multiply (GEMM) of two Array objects
get_active_backend: Get current active backend
get_available_backends: Get the available backends
get_backend_count: Get the available backend count
get_default_random_engine: Get default random engine
get_device: Get the current active device id
get_last_error: Fetch last error description as String
get_mem_step_size: Get the minimum memory chunk size
get_revision: Get ArrayFire Revision (commit) information of the library.
get_seed: Get the seed of random number generator
get_size: Get size, in bytes, of the arrayfire native type
get_version: Get ArrayFire Version Number
gradient: Calculate the gradients
gray2rgb: Grayscale to Color(RGB) conversion
gt: Perform greater than comparison operation
hamming_matcher: Hamming feature matcher
handle_error_general: Default error handling callback provided by ArrayFire crate
harris: Harris corner detector.
hist_equal: Histogram Equalization
histogram: Compute Histogram of an Array
homography: Homography estimation
hsv2rgb: HSV to RGB color space conversion
hypot: Compute length of hypotenuse of two Arrays
identity: Create an identity array with 1’s in diagonal
ifft: Inverse fast fourier transform for 1d signals
ifft2: Inverse fast fourier transform for 2d signals
ifft3: Inverse fast fourier transform for 3d signals
ifft2_inplace: In place 2d dimensional inverse fast fourier transform
ifft3_inplace: In place 3d dimensional inverse fast fourier transform
ifft_inplace: In place 1d dimensional inverse fast fourier transform
iir: Infinite impulse response filter
imag: Extract imaginary values from a complex Array
imax: Find maximum value along given dimension and their corresponding indices
imax_all: Find maximum and it’s index in the whole Array
imin: Find minimum value along given dimension and their corresponding indices
imin_all: Find minimum and it’s index in the whole Array
index: Indexes the input Array using seqs Sequences
index_gen: Index an Array using any combination of Array’s and Sequence’s
info: Print library meta-info
info_string: Return library meta-info as String
init: Initialize ArrayFire library
inverse: Compute inverse of a matrix
inverse_deconv: Inverse deconvolution
iota: Create a range of values
is_double_available: Check if a device has double support
is_eval_manual: Get eval flag value
is_half_available: Check if a device has half support
is_imageio_available: Function to check if Image I/O is available
is_lapack_available: Function to check if lapack support is available
isinf: Check if values are infinity
isnan: Check if values are NaN
iszero: Check if values are zero
iterative_deconv: Iterative Deconvolution
join: Join two arrays
join_many: Join multiple arrays
le: Perform less than equals comparison operation
lgamma: Compute the logarithm of absolute values of gamma function
load_image: Load Image into Array
load_image_native: Load Image into Array in it’s native type
locate: Locate the indices of non-zero elements.
log: Compute the natural logarithm
log2: Compute logarithm base 2
log1p: Compute the logarithm of input Array + 1
log10: Compute logarithm base 10
lookup: Lookup(hash) an Array using another Array
lower: Create lower triangular matrix
lt: Perform less than comparison operation
lu: Perform LU decomposition
lu_inplace: Perform inplace LU decomposition
match_template: Image matching
matmul: Matrix multiple of two Arrays
max: Find maximum among elements of given dimension
max_all: Find maximum among all values of the Array
max_by_key: Find maximum among values of similar keys along a given dimension
max_ragged: Max reduction along given axis as per ragged lengths provided
maxfilt: Box filter with maximum as box operation
maxof: Elementwise maximum operation of two Arrays
mean: Mean along specified dimension
mean_all: Compute mean of all data
mean_all_weighted: Compute weighted mean of all data
mean_shift: Meanshift Filter.
mean_weighted: Weighted mean along specified dimension
meanvar: Calculate mean and variance in single API call
medfilt: Median filter
medfilt1: One dimensional median filter on image
median: Find the median along a given dimension
median_all: Compute median of all data
min: Find minimum among elements of given dimension
min_all: Find minimum among all values of the Array
min_by_key: Find minimum among values of similar keys along a given dimension
minfilt: Box filter with minimum as box operation
minof: Elementwise minimum operation of two Arrays
moddims: Change the shape of the Array
modulo: Compute modulo of two Arrays
moments: Find Image moments
moments_all: Find Image moment for whole image
mul: Multiplication of two Arrays
nearest_neighbour: Nearest Neighbour.
neq: Elementwise not equals comparison of two Arrays
norm: Find the norm of a matrix
or: Elementwise logical or operation of two Arrays
orb: ORB feature descriptor
pad: Pad input Array along borders
pinverse: Psuedo Inverse of Matrix
pow: Computer power
pow2: Compute two raised to the power of value
print: Print data in the Array
print_gen: Generalized Array print function
print_mem_info: Print buffer details from the ArrayFire device manager
product: Compute product of elements along a given dimension
product_all: Product of all values of the Array
product_by_key: Find product of all values with similar keys along a given dimension
product_by_key_nan: Compute product of all values with similar keys along a given dimension
product_nan: Product of elements along specific dimension using user specified value instead of NAN values
product_nan_all: Product of all values using user provided value for NAN
qr: Perform QR decomposition
qr_inplace: Perform inplace QR decomposition
randn: Create random numbers from normal distribution
random_normal: Generate array of normal numbers using a random engine
random_uniform: Generate array of uniform numbers using a random engine
randu: Create random numbers from uniform distribution
range: Create a Range of values
rank: Find rank of a matrix
real: Extract real values from a complex Array
regions: Find blobs in given image.
register_error_handler: Register user provided error handler
rem: Compute remainder from two Arrays
reorderDeprecated: Reorder the array in specified order
reorder_v2: Reorder the array according to the new specified axes
replace: Inplace replace in Array based on a condition
replace_scalar: Inplace replace in Array based on a condition
resize: Resize an Image
rgb2gray: Color(RGB) to Grayscale conversion
rgb2hsv: RGB to HSV color space conversion
rgb2ycbcr: RGB to YCbCr colorspace converter.
root: Compute root
rotate: Rotate an Image
round: Round the values in an Array
row: Extract row_num row from input Array
rows: Get an Array with all rows from first to last in the input Array
rsqrt: Compute the reciprocal square root
sat: Summed area table of an Image
save_image: Save an Array to an image file
save_image_native: Save an Array without modifications to an image file
scale: Scale an Image
scan: Generalized scan
scan_by_key: Generalized scan by key
select: Element wise conditional operator for Arrays
selectl: Element wise conditional operator for Arrays
selectr: Element wise conditional operator for Arrays
set_backend: Toggle backends between cuda, opencl or cpu
set_col: Set col_num^th col in inout Array to a new Array new_col
set_cols: Set cols from first to last in inout Array with cols from Array new_cols
set_cublas_mode: Sets the cuBLAS math mode for the internal handle.
set_default_random_engine_type: Set the random engine type for default random number generator
set_device: Set active device
set_fft_plan_cache_size: Set fft plan cache size
set_intersect: Find intersection of two sets
set_manual_eval: Set eval flag value
set_mem_step_size: Set the minimum memory chunk size
set_row: Set row_num^th row in inout Array to a new Array new_row
set_rows: Set rows from first to last in inout Array with rows from Array new_rows
set_seed: Set seed for random number generation
set_slice: Set slice slice_num in inout Array to a new Array new_slice
set_slices: Set first to last slices of inout Array to a new Array new_slices
set_union: Find union of two sets
set_unique: Find unique values from a Set
shift: “Circular shift of values along specified dimension
shiftl: Compute left shift
shiftr: Compute right shift
sigmoid: Compute sigmoid function
sign: Computes the sign of input Array values
sin: Compute sin
sinh: Compute sinh
skew: Skew an image
slice: Get slice_num^th slice from input Array
slices: Get slices from first to last in input Array
sobel: Sobel Operator
solve: Solve a system of equations
solve_lu: Solve a system of equations
sort: Sort the values in input Arrays
sort_by_key: Sort the values in input Arrays
sort_index: Sort the values in input Arrays
sparse: Create sprase matrix from arrays
sparse_convert_to: Convert between sparse formats
sparse_from_dense: Convert dense array to sparse array
sparse_from_host: Create sprase matrix from data on host memory
sparse_get_col_indices: Get cololumn indices Array
sparse_get_format: Get sparse format
sparse_get_info: Get sparse Array information
sparse_get_nnz: Get number of non-zero elements in sparse array
sparse_get_row_indices: Get row indices Array
sparse_get_values: Get values of sparse Array
sparse_to_dense: Convert sparse array to dense array
sqrt: Compute the square root
stdevDeprecated: Standard deviation along specified axis
stdev_all: Compute standard deviation of all data
stdev_all_v2: Compute standard deviation of all data
stdev_v2: Standard deviation along given axis
sub: Subtraction of two Arrays
sum: Sum elements along a given dimension
sum_all: Sum all values of the Array
sum_by_key: Find sum of all values with similar keys along a given dimension
sum_by_key_nan: Compute sum of all values with similar keys along a given dimension
sum_nan: Sum along specific dimension using user specified value instead of NAN values
sum_nan_all: Sum all values using user provided value for NAN
susan: SUSAN corner detector.
svd: Perform Singular Value Decomposition
svd_inplace: Perform Singular Value Decomposition inplace
sync: Sync all operations on given device
tan: Compute tan
tanh: Compute tanh
tgamma: Compute gamma function
tile: Tile the input array along specified dimension
topk: Find top k elements along a given dimension
transform: Transform(Affine) an Image
transform_coords: Transform input coordinates
translate: Translate an Image
transpose: Transpose of a matrix.
transpose_inplace: Inplace transpose of a matrix.
trunc: Truncate the values in an Array
unwrap: Generate an array with image windows as columns
upper: Create upper triangular matrix
varDeprecated: Compute Variance along a specific dimension
var_allDeprecated: Compute Variance of all elements
var_all_v2: Compute Variance of all elements
var_all_weighted: Compute weighted variance of all data
var_v2: Compute Variance along a specific dimension
var_weighted: Weight variance along specified dimension
wrap: Converts unwrapped image to an image
ycbcr2rgb: YCbCr to RGB colorspace converter.

Type Aliases§

ErrorCallback: Signature of error handling callback function
af_array: ArrayFire FFI Type alias for af_array
af_event: ArrayFire FFI Type alias for af_event
af_features: ArrayFire FFI Type alias for af_features
af_index_t: ArrayFire FFI Type alias for af_indexers_t
af_random_engine: ArrayFire FFI Type alias for af_random_engine
af_window: ArrayFire FFI Type alias for af_window
c32: Short type alias for Complex single precision type
c64: Short type alias for Complex double precision type
dim_t: ArrayFire FFI Type alias for libc’s signed long long
u64_t: ArrayFire FFI Type alias for libc’s unsigned long long
void_ptr: ArrayFire FFI Type alias for libc’s void*

Crate arrayfireCopy item path