Visible to Intel only — GUID: GUID-A9E5DDF9-1382-4E19-81B8-3713058AE2FF
basic_statistics_dense_batch.cpp
basic_statistics_dense_online.cpp
column_accessor_homogen.cpp
cor_dense_batch.cpp
cor_dense_online.cpp
cov_dense_batch.cpp
cov_dense_biased_batch.cpp
cov_dense_biased_online.cpp
cov_dense_online.cpp
csr_accessor.cpp
csr_table.cpp
dbscan_brute_force_batch.cpp
df_cls_hist_batch.cpp
df_cls_hist_batch_random.cpp
df_cls_traverse_model.cpp
df_reg_hist_batch.cpp
df_reg_hist_batch_random.cpp
df_reg_traverse_model.cpp
heterogen_table.cpp
homogen_table.cpp
kmeans_init_dense.cpp
kmeans_lloyd_dense_batch.cpp
knn_cls_brute_force_dense_batch.cpp
knn_reg_brute_force_dense_batch.cpp
knn_search_brute_force_dense_batch.cpp
linear_kernel_dense_batch.cpp
linear_regression_dense_batch.cpp
linear_regression_dense_online.cpp
logistic_regression_dense_batch.cpp
pca_cor_dense_batch.cpp
pca_cor_dense_online.cpp
pca_cov_dense_batch.cpp
pca_cov_dense_online.cpp
pca_precomputed_cor_dense_batch.cpp
pca_precomputed_cov_dense_batch.cpp
pca_svd_dense_batch.cpp
rbf_kernel_dense_batch.cpp
svm_two_class_thunder_dense_batch.cpp
basic_statistics_dense_batch.cpp
basic_statistics_dense_online.cpp
column_accessor_homogen.cpp
connected_components_batch.cpp
cor_dense_batch.cpp
cor_dense_online.cpp
cov_dense_batch.cpp
cov_dense_biased_batch.cpp
cov_dense_biased_online.cpp
cov_dense_online.cpp
csr_accessor.cpp
csr_table.cpp
dbscan_brute_force_batch.cpp
df_cls_dense_batch.cpp
df_reg_dense_batch.cpp
directed_graph.cpp
graph_service_functions.cpp
heterogen_table.cpp
homogen_table.cpp
jaccard_batch.cpp
jaccard_batch_app.cpp
kmeans_init_dense.cpp
kmeans_lloyd_dense_batch.cpp
knn_cls_brute_force_dense_batch.cpp
knn_cls_kd_tree_dense_batch.cpp
knn_search_brute_force_dense_batch.cpp
linear_kernel_dense_batch.cpp
linear_regression_dense_batch.cpp
linear_regression_dense_online.cpp
logloss_dense_batch.cpp
louvain_batch.cpp
pca_cor_dense_batch.cpp
pca_cor_dense_online.cpp
pca_cov_dense_batch.cpp
pca_cov_dense_online.cpp
pca_precomputed_dense_batch.cpp
pca_svd_dense_batch.cpp
pca_svd_dense_online.cpp
polynomial_kernel_dense_batch.cpp
rbf_kernel_dense_batch.cpp
shortest_paths_batch.cpp
sigmoid_kernel_dense_batch.cpp
subgraph_isomorphism_batch.cpp
svm_multi_class_thunder_csr_batch.cpp
svm_multi_class_thunder_dense_batch.cpp
svm_nu_cls_thunder_csr_batch.cpp
svm_nu_cls_thunder_dense_batch.cpp
svm_nu_reg_thunder_csr_batch.cpp
svm_nu_reg_thunder_dense_batch.cpp
svm_reg_thunder_csr_batch.cpp
svm_reg_thunder_dense_batch.cpp
svm_two_class_smo_csr_batch.cpp
svm_two_class_smo_dense_batch.cpp
svm_two_class_thunder_csr_batch.cpp
svm_two_class_thunder_dense_batch.cpp
triangle_counting_batch.cpp
K-Means Clustering
Density-Based Spatial Clustering of Applications with Noise
Correlation and Variance-Covariance Matrices
Principal Component Analysis
Principal Components Analysis Transform
Singular Value Decomposition
Association Rules
Kernel Functions
Expectation-Maximization
Cholesky Decomposition
QR Decomposition
Outlier Detection
Distance Matrix
Distributions
Engines
Moments of Low Order
Quantile
Quality Metrics
Sorting
Normalization
Optimization Solvers
Decision Forest
Decision Trees
Gradient Boosted Trees
Stump
Linear and Ridge Regressions
LASSO and Elastic Net Regressions
k-Nearest Neighbors (kNN) Classifier
Implicit Alternating Least Squares
Logistic Regression
Naïve Bayes Classifier
Support Vector Machine Classifier
Multi-class Classifier
Boosting
Visible to Intel only — GUID: GUID-A9E5DDF9-1382-4E19-81B8-3713058AE2FF
Array
Refer to Developer Guide: Array.
Programming interface
All types and functions in this section are declared in the oneapi::dal namespace and be available via inclusion of the oneapi/dal/array.hpp header file.
All the array class methods can be divided into several groups:
Constructors that are used to create an array from external, mutable or immutable memory.
Constructors and assignment operators that are used to create an array that shares its data with another one.
The group of reset() methods that are used to re-assign an array to another external memory block.
The group of reset() methods that are used to re-assign an array to an internally allocated memory block.
The methods that are used to access the data.
Static methods that provide simplified ways to create an array either from external memory or by allocating it within a new object.
template<typenameT>classarray
- Template Parameters
-
T – The type of the memory block elements within the array. T can represent any type.
Public Static Methods
staticarray<T>empty(std::int64_tcount)
Allocates a new memory block for mutable data, does not initialize it, creates a new array instance by passing a pointer to the memory block. The array owns the memory block (for details, see data ownership requirements).
- Parameters
-
count – The number of elements of type Data to allocate memory for.
- Preconditions
-
count > 0
template<typenameK>staticarray<T>full(std::int64_tcount, K&&element)
Allocates a new memory block for mutable data, fills it with a scalar value, creates a new array instance by passing a pointer to the memory block. The array owns the memory block (for details, see data ownership requirements).
- Parameters
-
count – The number of elements of type T to allocate memory for.
element – The value that is used to fill a memory block.
- Preconditions
-
count > 0
Elements of type T are constructible from the Element type.
staticarray<T>zeros(std::int64_tcount)
Allocates a new memory block on mutable data, fills it with zeros, creates a new array instance by passing a pointer to the memory block. The array owns the memory block (for details, see data ownership requirements).
- Parameters
-
count – The number of elements of type Data to allocate memory for.
- Preconditions
-
count > 0
template<typenameY>staticarray<T>wrap(Y*data, std::int64_tcount)
Creates a new array instance by passing the pointer to externally-allocated memory block for mutable data. It is the responsibility of the calling application to free the memory block as the array does not free it when the reference count is zero.
- Parameters
-
data – The pointer to externally-allocated memory block.
count – The number of elements of type Data in the memory block.
Constructors
array()
Creates a new instance of the class without memory allocation: mutable_data and data pointers should be set to nullptr, count should be zero; the pointer to the ownership structure should be set to nullptr.
array(constarray<T>&other)
Creates a new array instance that shares an ownership with other on its memory block.
array(array<T>&&other)
Moves data, mutable_data pointers, count, and pointer to the ownership structure in other to the new array instance.
template<typenameDeleter>array(T*data, std::int64_tcount, Deleter&&deleter)
Creates a new array instance which owns a memory block of externally-allocated mutable data. The ownership structure is created for a block, the input deleter is assigned to it.
- Template Parameters
-
Deleter – The type of a deleter used to free the Data. The deleter provides void operator()(Data*) member function.
- Parameters
-
data – The pointer to externally-allocated memory block.
count – The number of elements of type Data in the memory block.
deleter – The object used to free Data.
template<typenameConstDeleter>array(constT*data, std::int64_tcount, ConstDeleter&&deleter)
Creates a new array instance which owns a memory block of externally-allocated immutable data. The ownership structure is created for a block, the input deleter is assigned to it.
- Template Parameters
-
ConstDeleter – The type of a deleter used to free the Data. The deleter implements void operator()(const Data*) member function.
- Parameters
-
data – The pointer to externally-allocated memory block.
count – The number of elements of type Data in the Data.
deleter – The object used to free Data.
array(conststd::shared_ptr<T>&data, std::int64_tcount)
Creates a new array instance that shares ownership with the user-provided shared pointer.
- Parameters
-
data – The shared pointer to externally-allocated memory block.
count – The number of elements of type Data in the memory block.
array(conststd::shared_ptr<constT>&data, std::int64_tcount)
Creates a new array instance that shares ownership with the user-provided shared pointer.
- Parameters
-
data – The shared pointer to externally-allocated memory block.
count – The number of elements of type Data in the memory block.
template<typenameY,typenameK>array(constarray<Y>&ref, K*data, std::int64_tcount)
An aliasing constructor: creates a new array instance that stores Data pointer, assigns the pointer to the ownership structure of ref to the new instance. Array returns Data pointer as its mutable or immutable block depending on the Data type.
- Template Parameters
-
Y – The type of elements in the referenced array.
K – Either T or
type.
- Parameters
-
ref – The array which shares ownership structure with created one.
data – Mutable or immutable unmanaged pointer hold by created array.
count – The number of elements of type T in the Data.
array(impl_t*impl)
Creates array from impl.
Public Methods
array<T>operator=(constarray<T>&other)
Replaces the data, mutable_data pointers, count, and pointer to the ownership structure in the array instance by the values in other.
array<T>operator=(array<T>&&other)
Swaps the values of data, mutable_data pointers, count, and pointer to the ownership structure in the array instance and other.
T*get_mutable_data()const
The pointer to the memory block holding mutable data.
- Preconditions
-
has_mutable_data() == true, othewise throws domain_error
constT*get_data()constnoexcept
The pointer to the memory block holding immutable data.
boolhas_mutable_data()constnoexcept
Returns whether array contains mutable_data or not.
array&need_mutable_data()
Returns mutable_data, if array contains it. Otherwise, allocates a memory block for mutable data and fills it with the data stored at data. Creates the ownership structure for allocated memory block and stores the pointer.
- Postconditions
-
has_mutable_data() == true
std::int64_tget_count()constnoexcept
The number of elements of type T in a memory block.
std::int64_tget_size()constnoexcept
The size of memory block in bytes.
voidreset()
Resets ownership structure pointer to nullptr, sets count to zero, data and mutable_data to nullptr.
voidreset(std::int64_tcount)
Allocates a new memory block for mutable data, does not initialize it, creates ownership structure for this block, assigns the structure inside the array. The array owns allocated memory block.
- Parameters
-
count – The number of elements of type Data to allocate memory for.
template<typenameDeleter>voidreset(T*data, std::int64_tcount, Deleter&&deleter)
Creates the ownership structure for memory block of externally-allocated mutable data, assigns input deleter object to it, sets data and mutable_data pointers to this block.
- Template Parameters
-
Deleter – The type of a deleter used to free the Data. The deleter implements void operator()(Data*) member function.
- Parameters
-
data – The mutable memory block pointer to be assigned inside the array.
count – The number of elements of type Data into the block.
deleter – The object used to free Data.
template<typenameConstDeleter>voidreset(constT*data, std::int64_tcount, ConstDeleter&&deleter)
Creates the ownership structure for memory block of externally-allocated immutable data, assigns input deleter object to it, sets data pointer to this block.
- Template Parameters
-
ConstDeleter – The type of a deleter used to free. The deleter implements void operator()(const Data*)` member function.
- Parameters
-
data – The immutable memory block pointer to be assigned inside the array.
count – The number of elements of type Data into the block.
deleter – The object used to free Data.
template<typenameY>voidreset(constarray<Y>&ref, T*data, std::int64_tcount)
Initializes data and mutable_data with data pointer, count with input count value, initializes the pointer to ownership structure with the one from ref. Array returns Data pointer as its mutable block.
- Template Parameters
-
Y – The type of elements in the referenced array.
- Parameters
-
ref – The array which is used to share ownership structure with current one.
data – Mutable unmanaged pointer to be assigned to the array.
count – The number of elements of type T in the Data.
template<typenameY>voidreset(constarray<Y>&ref, constT*data, std::int64_tcount)
Initializes data with data pointer, count with input count value, initializes the pointer to ownership structure with the one from ref. Array returns Data pointer as its immutable block.
- Template Parameters
-
Y – The type of elements in the referenced array.
- Parameters
-
ref – The array which is used to share ownership structure with current one.
data – Immutable unmanaged pointer to be assigned to the array.
count – The number of elements of type T in the Data.
constT&operator[](std::int64_tindex)constnoexcept
Provides a read-only access to the elements of array. Does not perform boundary checks.
array<T>get_slice(std::int64_tfirst, std::int64_tlast)const
Creates slice of this array.
Usage Example
The following listing provides a brief introduction to the array API and an example of basic usage scenario:
#include <sycl/sycl.hpp>
#include <iostream>
#include <string>
#include "oneapi/dal/array.hpp"
using namespace oneapi;
void print_property(const std::string& description, const auto& property) {
std::cout << description << ": " << property << std::endl;
}
int main() {
sycl::queue queue { sycl::default_selector() };
constexpr std::int64_t data_count = 4;
const float data[] = { 1.0f, 2.0f, 3.0f, 4.0f };
// Creating an array from immutable user-defined memory
auto arr_data = dal::array<float>::wrap(data, data_count);
// Creating an array from internally allocated memory filled by ones
auto arr_ones = dal::array<float>::full(queue, data_count, 1.0f);
print_property("Is arr_data mutable", arr_data.has_mutable_data()); // false
print_property("Is arr_ones mutable", arr_ones.has_mutable_data()); // true
// Creating new array from arr_data without data copy - they share ownership information.
dal::array<float> arr_mdata = arr_data;
print_property("arr_mdata elements count", arr_mdata.get_count()); // equal to data_count
print_property("Is arr_mdata mutable", arr_mdata.has_mutable_data()); // false
/// Copying data inside arr_mdata to new mutable memory block.
/// arr_data still refers to the original data pointer.
arr_mdata.need_mutable_data(queue);
print_property("Is arr_data mutable", arr_data.has_mutable_data()); // false
print_property("Is arr_mdata mutable", arr_mdata.has_mutable_data()); // true
queue.submit([&](sycl::handler& cgh){
auto mdata = arr_mdata.get_mutable_data();
auto cones = arr_ones.get_data();
cgh.parallel_for<class array_addition>(sycl::range<1>(data_count), [=](sycl::id<1> idx) {
mdata[idx[0]] += cones[idx[0]];
});
}).wait();
std::cout << "arr_mdata values: ";
for(std::int64_t i = 0; i < arr_mdata.get_count(); i++) {
std::cout << arr_mdata[i] << ", ";
}
std::cout << std::endl;
return 0;
}