[−][src]Crate arrayfire
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 |
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 |
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 |
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 |
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 | Perform signal interpolation for 2d signals |
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 |
color_space | Color space conversion |
cols | Get all cols from |
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 |
convolve2_gradient_nn | Backward pass gradient of 2D convolution |
convolve2_nn | Convolution Integral for two dimensional data |
convolve2_sep | Separable convolution for 2d signals |
convolve3 | 3d convolution |
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 |
cov | Deprecated 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 |
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 |
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 | Fast fourier transform for 3d signals |
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 |
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 |
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 |
ifft2_inplace | In place 2d dimensional inverse fast fourier transform |
ifft3 | Inverse fast fourier transform for 3d signals |
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 |
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 |
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 |
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 |
log1p | Compute the logarithm of input Array + 1 |
log2 | Compute logarithm base 2 |
log10 | Compute logarithm base 10 |
lookup | Lookup(hash) an Array using another Array |
lower | Create lower triangular matrix |
lt | Perform |
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 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 |
product_nan_all | Product of all values using user provided value for |
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 |
reorder | Deprecated 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 |
rows | Get an Array with all rows from |
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 |
set_cols | Set cols from |
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 |
set_rows | Set rows from |
set_seed | Set seed for random number generation |
set_slice | Set slice |
set_slices | Set |
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 |
slices | Get slices from |
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 |
stdev | Deprecated 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 |
sum_nan_all | Sum all values using user provided value for |
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 |
var | Deprecated Compute Variance along a specific dimension |
var_all | Deprecated 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 Definitions
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* |