China
Africa
Explore chapters and articles related to this topic
Calling C
Published in David E. Hiebeler, R and MATLAB®, 2018
Code similar to that shown below was used to measure the time needed to repeatedly permute a set of values in six different ways:Using .C() to call the Cshuffle C function directly.Using the wrapper function shuffle2.Using the wrapper function shuffle.Using the sample function.Using .Call() with the Cshufflecall C function (see Page 186).Using .C() to call the Cshuffle2 C function, which uses 's random-number generator (see Page 183).
Cloud Computing: The Flexible Future
Published in Hrishikesh Venkatarman, Ramona Trestian, 5G Radio Access Networks: Centralized RAN, Cloud-RAN, and Virtualization of Small Cells, 2017
Joanna Kusznier, Xuan Thuy Dang, Manzoor Ahmed Khan
OpenNebula is a project originated in the European academic world. It is an open-source cloud computing toolkit for managing heterogeneous distributed data center infrastructures. The OpenNebula platform manages a data center’s virtual infrastructure to build private, public, and hybrid implementations of infrastructure as a service [34]. Its important characteristic is that OpenNebula does not have any specific infrastructural requirements; therefore, it is easier to fit in the existing environment. It is based on the idea of OpenNebula as a purely private cloud, in which users actually log into the head node to access cloud functions. This interface is a wrapper around an XML remote procedure call (XML-RPC) interface, which can also be used directly. Sempolinski [35] note that a front-end interface, such as the Elastic Compute Cloud (EC2), can be appended to this default configuration.
Service Deployment
Published in Krzysztof W. Kolodziej, Johan Hjelm, Local Positioning Systems, 2017
Krzysztof W. Kolodziej, Johan Hjelm
In order to make systems available on other platforms, the equivalent of the system’s specific WRAPI application class will need to be reimplemented for that system. The wrapper class exposes an interface to WRAPI. As long as that interface remains unchanged, WRAPI could be replaced or a similar API developed for different systems, without any impact on other portions of the application. In order to access the hardware, operating system-specific system calls need to be made, which makes the API platform dependent. These system calls, though similar in concept, are different in their usage on different operating systems.
Feature Selection Empowered by Self-Inertia Weight Adaptive Particle Swarm Optimization for Text Classification
Published in Applied Artificial Intelligence, 2022
Muhammad Asif, Arfan Ali Nagra, Maaz Bin Ahmad, Khalid Masood
The steps involved in selecting feature sub-set can be divided into wrappers, filters, and embedded approaches. The filter’s task is to separate FS from the learning algorithm and choose a sub-set that shouldn’t depend on any particular learning algorithm (Lee et al. 2019). The evaluation method is used in the wrapper approach to select the feature sub-set. It is based on an exact learning algorithm to be used in the next step. During the evaluation process, the efficiency and sustainability of the sub-sets are checked to find the better one. The comparison of the sub-set with the prior best particle is also part of it. A stopping condition is checked at each iteration to find whether the FS should continue or stop. The wrapper function is considered a better solution generator because it is complex and can break into many features. If the FS and learning algorithm are interleaved, then the FS procedure falls in the domain of the embedded function (Wu et al. 2013).
Systematic comparison of path planning algorithms using PathBench
Published in Advanced Robotics, 2022
Hao-Ya Hsueh, Alexandru-Iosif Toma, Hussein Ali Jaafar, Edward Stow, Riku Murai, Paul H. J. Kelly, Sajad Saeedi
An overview of the architecture of PathBench is shown in Figure 2. PathBench is composed of four main components: simulator, generator, trainer, and analyzer where infrastructures are created to link the four main components with other parts of the framework to provide general service libraries and utilities. The simulator is responsible for environment interactions and algorithm visualization. It provides custom collision detection systems and a graphics framework for rendering the internal state of the algorithms. The generator is responsible for generating and labelling the training data used to train the ML models. The trainer is a class wrapper over the third-party machine learning libraries. It provides a generic training pipeline based on the holdout method and standardized access to the training data. Finally, the analyzer manages the statistical measures used in the practical assessment of the algorithms. Custom metrics can be defined, as well as graphical displays for visual comparisons. PathBench has been written in Python and uses PyTorch [58] for ML.
spc4sts: Statistical process control for stochastic textured surfaces in R
Published in Journal of Quality Technology, 2021
The results in this article are obtained using 64-bit 3.6.1 with the spc4sts 0.5.2 package and a computer with an Intel® Core™ i9-9900K CPU at 3.6 GHz and 64 GB of RAM on Windows 10. This package makes use of the rpart 4.1-13 package to fit rpart trees. The stationary test in the surfacemodel function of spc4sts is performed by calling stationaryTest of spc4sts, which is a wrapper function of the TOS2D function in the LS2Wstat 2.1-1 package of Taylor and Nunes (2018). The showNb function of spc4sts makes use of the grid.table function in the gridExtra 2.3 package of Auguie (2017). The textile images in Section 4 are from the textile 0.1.2 package. itself and these packages are available from the Comprehensive Archive Network (CRAN) at https://CRAN.R-project.org/. The resize function used to stretch images is in the EBImage 4.26.0 package of Pau et al. (2010), which is available from Bioconductor at https://bioconductor.org/. The textile2 package is from Mendeley data at http://dx.doi.org/10.17632/wy3pndgpcx.1.