intel Get Started with oneAPI oneAPI Data Analytics Library User Guide

: June 9, 2024
: Intel

Table of Contents

Get Started with Intel® one API one API Data Analytics Library
References
Read User Manual Online (PDF format)
Download This Manual (PDF format)

Get Started with Intel®
one API one API **Data Analytics Library**

Get Started with Intel® one API one API Data Analytics Library

Intel® oneAPI Data Analytics Library (oneDAL) is a library that helps speed up big data analysis by providing highly optimized algorithmic building blocks for all stages of data analytics (preprocessing, transformation, analysis, modeling, validation, and decision making) in batch, online, and distributed processing modes of computation. For general information about oneDAL, visit oneDAL official page.

Before You Begin
oneDAL is located in /dal directory where is the directory in which Intel® oneAPI Base Toolkit was installed.
The current version of oneDAL with SYCL support is available for Linux and Windows 64-bit operating systems. The prebuilt oneDAL libraries can be found in the /dal//resist directory.
To learn about the system requirements and the dependencies needed to build examples, refer to the System Requirements page.

End-to-end Example
Below you can find a typical usage workflow for a oneDAL algorithm on GPU. The example is provided for Principal Component Analysis algorithm (PCA).

The following steps depict how to:

Read the data from CSV file
Run the training and inference operations for PCA
Access intermediate results obtained at the training stage

Include the following header that makes all oneDAL declarations available.
Create a SYCL* queue with the desired device selector. In this case, GPU selector is used:
const auto queue = sycl::queue{ sycl::gpu_selector{} };
Since all oneDAL declarations are in the oneapi::dal namespace, import all declarations from the oneapi namespace to use dal instead of oneapi::dal for brevity: using namespace oneapi;
Use CSV data source to read the data from the CSV file into a table:
const auto data = dal::read(queue, dal::csv::data source{“data.csv”});
Create a PCA descriptor, configure its parameters, and run the training algorithm on the data loaded from CSV.
Print the learned eigenvectors:
Use the trained model for inference to reduce dimensionality of the data:

Build and Run Examples
Perform the following steps to build and run examples demonstrating the basic usage scenarios of one DAL with SYCL support. Go to

/dal/ and then set up an environment as shown in the example below:

NOTE All content below that starts with # is considered a comment and should not be run with the code.

Set up the required environment for one DAL (variables such as CPATH, LIBRARY_PATH, and LD_LIBRARY_PATH):
• On Linux, there are two possible ways to set up the required environment: via vars.sh script or via module files.
• Setting up one DAL environment via vars.sh script Run the following command:
source ./env/vars.sh
• Setting up one DAL environment via module files
Initialize1. modules: source $MODULESHOME/infit/bash
NOTE Refer to Environment Modules documentation for details.
Provide2. modules with a path to the module files directory: module
use ./module files
Run the3.module:
module load dal
• On Windows, run the following command:
/env/vars.bat
Copy ./examples/neap/dip to a writable directory if necessary (since it creates temporary files):
cp r ./examples/oneapi/dpc ${WRITABLE_DIR}
Set up the compiler environment for Intel® one API DPC++/C++ Compiler. See Get Started with Intel® oneAPI DPC++/C++ Compiler for details.
Build and run the examples that show how to use one DAL with SYCL support:
NOTE You need to have write permissions to the examples folder to build examples, and execute permissions to run them. Otherwise, you need to copy examples/oneapi/dpc and examples/ oneapi/data folders to the directory with right permissions. These two folders must be retained in the same directory level relative to each other.
• On Linux:
• On Windows:
To see all available parameters of the build procedure, type make on Linux or unmake on Windows.
The resulting example binaries and log files are written into the _results directory.
NOTE You should run the examples from examples/oneapi/dpc folder, not from _results folder. Most examples require data to be stored in examples/oneapi/data folder and to have a relative link to it started from examples/oneapi/dip folder.
You can build traditional C++ examples located in examples/oneapi/cup folder in a similar way.

Compile and build applications with pkg-config
The pkg-config tool is a widely used tool for building software with dependencies. Intel® one API Data Analytics Library provides files with pug- config metadata for compiling and linking an application to the library.

Set up the environment
To use pug-config, build the library and then set up the environment using vars.sh or vars.bat scripts:

On Linux: source ./env/vars.sh
On Windows: /env/vars.bat

Choose a metadata file
The metadata files provided by one DAL cover only host device configuration on 64-bit Linux, macOS, or Windows operating system for C++.
Choose the metadata file based on one DAL threading mode and linking method you use:

one DAL pug-config metadata files

Single-threaded (non-threaded)	Multi-threaded (internally threaded)
Static linking	dal-static-sequential-host
Dynamic linking	dal-dynamic-sequential-host

Compile a program using peg-config
To compile a test.cpp program with one DAL and peg-config, provide the name of the one DAL pug-config metadata file as an input parameter. For example:

On Linux or macOS:
ice test.cpp pug-config –flags –libs dal-dynamic-threading-host
On Windows:
for /F “delimbs=,” %i in (‘pug-config –flags –libs dal-dynamic-threading-host) do icl test.cpp %i
A sample code for svm_two_class_thunder_dense_batch example with SYCL support. Run the following from the examples/oneapi/cpp directory:
On Linux or macOS:
icc -I source/ source/svm/svm_two_class_thunder_dense_batch.cpp icc test.cpp pkg-config –cflags –libs dal-dynamic-threading-host
On Windows:
for /F “delis=,” %i in (‘peg-config –flags –libs dal-dynamic-threading-host) do ill -I source/ ill svm_two_class_thunder_dense_batch.cpp %i

Find More

Document	Description
Developer Guide and Reference	Refer to one DAL Developer Guide and Reference

for
detailed information about implemented algorithms.
System Requirements| Check system requirements before you install Intel® one API
Data Analytics Library.
Release Notes| Refer to release notes for Intel® one API Data Analytics
Library to learn about new updates in the latest release.
Code Samples| Learn how to use one DAL with daal4py, a Python* API.
one DAL Specification| Learn about requirements for implementations of one API
Data Analytics Library.

Notices and Disclaimers
Intel technologies may require enabled hardware, software or service activation.
No product or component can be absolutely secure.
Your costs and results may vary.

© Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others.
No license (express or implied, by estoppel or otherwise) to any intellectual property rights is granted by this document.
The products described may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request.
Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.