Background: Classification algorithms assign observations to groups based on patterns in data. The machine-learning community have developed myriad classification algorithms, which are used in diverse life science res...
详细信息
Background: Classification algorithms assign observations to groups based on patterns in data. The machine-learning community have developed myriad classification algorithms, which are used in diverse life science research domains. algorithm choice can affect classification accuracy dramatically, so it is crucial that researchers optimize the choice of which algorithm(s) to apply in a given research domain on the basis of empirical evidence. In benchmark studies, multiple algorithms are applied to multiple datasets, and the researcher examines overall trends. In addition, the researcher may evaluate multiple hyperparameter combinations for each algorithm and use feature selection to reduce data dimensionality. Although software implementations of classification algorithms are widely available, robust benchmark comparisons are difficult to perform when researchers wish to compare algorithms that span multiple software packages. Programming interfaces, data formats, and evaluation procedures differ across software packages;and dependency conflicts may arise during installation. Findings: To address these challenges, we created ShinyLearner, an open-source project for integrating machine-learning packages into software containers. ShinyLearner provides a uniform interface for performing classification, irrespective of the library that implements each algorithm, thus facilitating benchmark comparisons. In addition, ShinyLearner enables researchers to optimize hyperparameters and select features via nested cross-validation;it tracks all nested operations and generates output files that make these steps transparent. ShinyLearner includes a Web interface to help users more easily construct the commands necessary to perform benchmark comparisons. ShinyLearner is freely available at https://***/srp33/ShinyLearner. Conclusions: This software is a resource to researchers who wish to benchmark multiple classification or feature-selection algorithms on a given dataset. We
This paper introduces a method of the H.264 inverse quantization implementation based on the ADSP-BF533 combining with specific quantization algorithm in H.264. And we describe the procedure of algorithm optimization ...
详细信息
This paper introduces a method of the H.264 inverse quantization implementation based on the ADSP-BF533 combining with specific quantization algorithm in H.264. And we describe the procedure of algorithm optimization using Blackfin system and the feature of BF533 instructions. By adjusting the searching table and structure of sentences, based on parallel and vector instructions, the algorithm improves the efficiency of inverse quantization. It is proved that the executive efficiency of the inverse quantization has been increased significantly compared with the inverse quantization in JM.
In this paper, a genetic algorithm optimization is applied to a meanderline in order to achieve a low-profile broadband polarization converting surface, also known as a twist reflector. A novel, single-layer, genetica...
详细信息
ISBN:
(纸本)9781424420414
In this paper, a genetic algorithm optimization is applied to a meanderline in order to achieve a low-profile broadband polarization converting surface, also known as a twist reflector. A novel, single-layer, genetically-engineered meanderline on a substrate of thickness t < 0.10λ{sub}o is presented. Modeled and measured results are in good agreement, demonstrating that polarization conversion occurs for 98% of power at a center frequency of 12.9 GHz with greater than 18% bandwidth.
In 2013, the proposal of China's "the Belt and Road" policy marked the birth of a new stage of international talent training. Under this macro background, the training goal of business and enterprise man...
详细信息
In 2013, the proposal of China's "the Belt and Road" policy marked the birth of a new stage of international talent training. Under this macro background, the training goal of business and enterprise management professionals in China's colleges should be oriented to high-tech application-oriented talents working in the front line, which is consistent with the demand of the international vision of building the "the Belt and Road". In addition, the construction of the "the Belt and Road" infrastructure also requires the service and support of colleges at the level of school enterprise integration. This manuscript takes the data mining algorithm as the starting point to deeply explore the purpose and training mode of the quality evaluation of the integration of industry and education under the background of the "the Belt and Road" and obtain a quantitative model. From the results of dynamic clustering experiments, it can be seen that when the three categories are clustered, the first category of sample A is the "school enterprise integrated talent training environment" with high scores, of which samples 2 and 6 are more similar. The second type of sample B is that the evaluation score of "school enterprise integration and cooperation effect" is slightly lower while the evaluation score of "school enterprise integration and cooperation process" is higher, and the fitting degree of samples 3 and 8 is better. The evaluation scores of the three types of sample C are "school enterprise integration enterprise quality' and "school enterprise integration cooperation effect", while the deviation degree of F value of 7 and 9 is smaller. Through cluster analysis and data fitting of different samples, it provides a new method and new idea for constructing a quantitative, whole process, multi-level and multi-level quality evaluation model of school enterprise integration of industrial and commercial enterprise management specialty in colleges.
暂无评论