CNN is one of the most successful deep learning methods
due to its remarkable performance on ImageNet large-scale
visual recognition competition. The success of CNN is due to
its capacity to learn hierarchical representation to describe the
image scene.
CNN is one of the most successful deep learning methods
due to its remarkable performance on ImageNet large-scale
visual recognition competition. The success of CNN is due to
its capacity to learn hierarchical representation to describe the
image scene. (2018-01-24, WINDOWS, 165KB, 下载2次)
对于各类时序信号，进行加入白噪声的eemd信号分解，此eemd为中央大学数据中心版本，亲测有效。有参数说明
For all kinds of time series signals, the EEMD signal is decomposed with white noise. Parameter description (2018-01-10, WINDOWS, 135KB, 下载3次)
%N是数据一共分多少类
%data是输入的不带分类标号的数据
%u是每一类的中心
%re是返回的带分类标号的数据
%N is the number of data in total
%data is an input data without a classified label
%u is the center of every category
%re is the returned data with classified labels (2017-11-01, WINDOWS, 1KB, 下载4次)
免疫算法是受生物免疫系统的启发而推出的一种新型的智能搜索算法，是一种确定性和随机性选择相结合并具有"勘探"与"开采"能力的启发式随机搜索算法。
算法主要的步骤:
(1)抗原识别与初始抗体产生。
(2)抗体评价
(3)免疫操作
免疫算法的特点:
(1)全局搜索能力
(2)多样性保持机制
(3)鲁棒性强
(4)并行分布式搜索机制
A new intelligent immune algorithm is inspired by the biological immune system and the introduction of the search algorithm is a deterministic and random selection are combined with "heuristic random search algorithm exploration" and "Exploitation" ability.
Main steps of the algorithm:
(1) antigen recognition and initial antibody production.
(2) antibody evaluation
(3) immunization
Characteristics of immune algorithm:
(1) global search capability
(2) diversity retention mechanism
(3) robust
(4) parallel distributed search mechanism (2017-10-05, WINDOWS, 27KB, 下载7次)
案例分析 全国研究生数学建模竞赛由教育部学位与研究生教育发展中心（以下简称学位中心）主办，是学位中心主办的“全国研究生创新实践系列活动”主题赛事之一。
matble The National Graduate Mathematical Contest is sponsored by the Ministry of Education Degree and Graduate Education Development Center (hereinafter referred to as the Degree Center) and is one of the thematic events of the "National Graduate Successful Practice Series" organized by the degree center. (2017-08-10, WINDOWS, 122KB, 下载1次)
实现K均值聚类，N为分类个数，data作为输入数据，re为输出带标签数据，u是最终聚类中心
To achieve K means clustering, N is the number of classification, data as input data, re as output tag data, and u is the final clustering center (2017-08-03, WINDOWS, 1KB, 下载5次)