Description: 本程序包含语音压缩和语音识别领域所需的LPCC，MFCC特征提取算法以及语音端点检测源码。在对语音数据进行特征提取前，可对语音数据进行16K到8K的降采样率处理，包含180阶FIR滤波器的频率压缩程序。-this program includes voice compression and voice recognition requirements in the area of the LPCC Features. MFCC feature extraction algorithm and voice endpoint detection source. In the voice data for feature extraction, voice data can be right for K to 16 K, the reduction of eight sampling rate, contains 180 bands FIR filter frequency compression process. Platform: |
Size: 39720 |
Author:王小飞 |
Hits:
Description: 一个acm算法题目 问题描述： 有一个黑盒子，可视为一个15×10的二维矩阵，里面有一些空隙，在二维矩阵中用0表示空隙，用1表示被占有。现有一个仪器，可以从4个方向对盒子进行分行透视：从左到右，从左下到右上，从下到上，从右下到左上。透视可得到该二维矩阵的该行1的个数。要求，还原此矩阵。 输入：4行，每一行分别代表从左到右，从左下到右上，从下到上，从右下到左上透视所得的数据 输出：该矩阵 使用的方法是回溯法，具体见程序-a acm algorithm topic Problem description : a black box, as a 15 x 10 two-dimensional matrix, there are some gaps in two-dimensional matrix using 0 indicates the gap, with a show that occupies. An existing equipment, from the direction of four branches box for perspective : from left to right, from the lower left to upper right, from bottom to top, from the lower right to upper left. Perspective availability of the two-dimensional matrix of the line a number. Demand reduction of this matrix. Input : 4 firms on behalf of every line from left to right, from the lower left to upper right, from bottom to top, from the upper left to lower right perspective on data obtained from the output : the matrix method is the use of backtracking, see the specific procedures Platform: |
Size: 5087 |
Author:陶锐 |
Hits:
Description: 一个acm算法题目 问题描述： 有一个黑盒子，可视为一个15×10的二维矩阵，里面有一些空隙，在二维矩阵中用0表示空隙，用1表示被占有。现有一个仪器，可以从4个方向对盒子进行分行透视：从左到右，从左下到右上，从下到上，从右下到左上。透视可得到该二维矩阵的该行1的个数。要求，还原此矩阵。 输入：4行，每一行分别代表从左到右，从左下到右上，从下到上，从右下到左上透视所得的数据 输出：该矩阵 使用的方法是回溯法，具体见程序-a acm algorithm topic Problem description : a black box, as a 15 x 10 two-dimensional matrix, there are some gaps in two-dimensional matrix using 0 indicates the gap, with a show that occupies. An existing equipment, from the direction of four branches box for perspective : from left to right, from the lower left to upper right, from bottom to top, from the lower right to upper left. Perspective availability of the two-dimensional matrix of the line a number. Demand reduction of this matrix. Input : 4 firms on behalf of every line from left to right, from the lower left to upper right, from bottom to top, from the upper left to lower right perspective on data obtained from the output : the matrix method is the use of backtracking, see the specific procedures Platform: |
Size: 5120 |
Author:陶锐 |
Hits:
Description: LWIP是TCP/IP协议栈的一种实现。LWIP的主要目的是减少存储器利用量和代码尺寸，使LWIP适合应用于小的、资源有限的处理器如嵌入式系统。为了减少处理器和存储器要求，lwIP可以通过不需任何数据拷贝的API进行裁减。
本文叙述了lwIP的设计与实现。叙述了协议实现及子系统中所使用的算法和数据结构如存储和缓冲管理系统。还包括LWIP API的参考手册和使用LWIP 的一些代码例子。-LWIP is TCP/IP Protocol Stack A. LWIP the main purpose is to reduce memory utilization and code size, LWIP suited for small, limited resources such as the processor embedded system. In order to reduce the processor and memory requirements, lwIP can copy data without any of the API for the reduction. This paper describes the lwIP Design and Implementation. Describes the protocol and subsystems used in the algorithm and data structure such as memory and buffer management system. LWIP API also includes a reference manual and use LWIP some code examples. Platform: |
Size: 693248 |
Author:tumbler |
Hits:
Description: EEPROM是一种可以进行电擦写，掉电后数据不丢失的存储器。它具有功耗
低、容量大、烧写与擦除速度快等优点，并且内部嵌入算法完成对芯片的操作，简化了软件的工作量，因而在数字信号处理系统中得到了广泛的应用。在DSP系统中通常在上电或复位时从外部加载并执行用户的程序代码，这个过程叫做自举加载。现要将程序空间的代码烧写入扩展的外部EEPROM中，以实现自举加载功能。
-EEPROM can be an electric erasable, power-fail without loss of data after the memory. It has low power consumption, large capacity, and the erasure of burning speed advantages, and the internal algorithm embedded chip to complete the operation, simplify the software workload, which in digital signal processing system, which is widely used. DSP system-- usually in the reduction or loaded from external users and the implementation of the code, This process is called bootstrap loading. Is the space should procedure code burned into the expansion of the external EEPROM, to achieve self-loading function held. Platform: |
Size: 4096 |
Author:深蓝 |
Hits:
Description: 本程序包含语音压缩和语音识别领域所需的LPCC，MFCC特征提取算法以及语音端点检测源码。在对语音数据进行特征提取前，可对语音数据进行16K到8K的降采样率处理，包含180阶FIR滤波器的频率压缩程序。-this program includes voice compression and voice recognition requirements in the area of the LPCC Features. MFCC feature extraction algorithm and voice endpoint detection source. In the voice data for feature extraction, voice data can be right for K to 16 K, the reduction of eight sampling rate, contains 180 bands FIR filter frequency compression process. Platform: |
Size: 39936 |
Author:王小飞 |
Hits:
Description: 一种通过自组织竞争学习网络实现数据降维和可视化的单层神经网络模型。用此算法可以把输入空间的多维映射到低维的（一维或者二维）的离散网络上，并将保持相同性质的输入数据在映射到低维空间时的拓扑一致性。iris以及letter两个数据集进行分类-A competitive learning through self-organizing network for data dimensionality reduction and visualization of single-layer neural network model. Using this algorithm can be multi-dimensional input space is mapped to the low-dimensional (one-dimensional or two-dimensional) discrete network, and will remain the same as the nature of the input data is mapped to the low-dimensional space of topological consistency. iris as well as the letter the two data sets to classify Platform: |
Size: 1024 |
Author:军军 |
Hits:
Description: pca算法，用于数据降维，注释非常详细清晰，-PCA algorithm, for data dimensionality reduction, clear and very detailed notes, Platform: |
Size: 1024 |
Author:zc |
Hits:
Description: 一种邻域保护嵌入算法，用于高维数据的降维处理-The protection of a neighborhood algorithm for high dimensional data, dimensionality reduction to deal with Platform: |
Size: 1024 |
Author:xinying |
Hits:
Description: pca算法的matlab实现 主成分分量分析可用于数据的降维和模式识别问题 -pca algorithm matlab component analysis to achieve the principal component can be used for data dimensionality reduction and pattern recognition problem Platform: |
Size: 1024 |
Author:薛睿 |
Hits:
Description: 支持IDE、SCSI硬盘。支持目前市面上所有硬盘品牌，本软件使用独特算法对硬盘进行快速、不破坏任何数据的检测，40G的7200转硬盘在雷鸟Xp1600 上实测速度为16分钟。最主要的是本软件支持任何测试 例：1 -99 、30 -80 、99 -1等。最大化的减少硬盘检测时间。所有坏道在本软件下都暴露无疑。-Support for IDE, SCSI hard drives. Support all hard disk brands currently on the market, this software uses a unique algorithm to drive fast, do not destroy any data detection, 40G 7200 RPM hard drives in Thunderbird Xp1600 on the measured speed of 16 minutes. Most important thing is the software support any testing cases: 1 -99 , 30 -80 , 99 -1 and so on. Maximize the reduction of the hard disk detection time. Under all the bad sectors are exposed to this software, no doubt. Platform: |
Size: 752640 |
Author:陈东 |
Hits:
Description: Kernel Entropy Component Analysis,KECA方法的作者R. Jenssen自己写的MATLAB代码，文章发表在2010年5月的IEEE TPAMI上面-Kernel Entropy Component Analysis, by R. Jenssen, published in IEEE TPAMI 2010.
We introduce kernel entropy component analysis (kernel ECA) as a new method for data transformation and dimensionality reduction. Kernel ECA reveals structure relating to the Renyi entropy of the input space data set, estimated via a kernel matrix using Parzen windowing. This is achieved by projections onto a subset of entropy preserving kernel principal component analysis (kernel PCA) axes. This subset does not need, in general, to correspond to the top eigenvalues of the kernel matrix, in contrast to the dimensionality reduction using kernel PCA. We show that kernel ECA may produce strikingly different transformed data sets compared to kernel PCA, with a distinct angle-based structure. A new spectral clustering algorithm utilizing this structure is developed with positive results. Furthermore, kernel ECA is shown to be an useful alternative for pattern denoising. Platform: |
Size: 3072 |
Author:johhnny |
Hits:
Description: Rose2里含有好多粗糙集的算法，可以实现数据预处理约简，求关联规则，求聚类和分类，非常实用。-Rose2 contains a lot of rough set algorithm, can achieve reduction of data preprocessing, find association rules, clustering and classification requirements, very practical. Platform: |
Size: 4145152 |
Author:曹盛文 |
Hits:
Description: 流形学习，局部线性嵌入式算法(LLE)，一种智能的算法去推测捕捉高维空间中所包含的低维特征。与适合于局部维数约减的聚类算法不同，LLE算法在单一的低维的全域坐标系统中表征采样空间，然而它并没有优化最小局域。通过对线性重构的局域对称的研究应用，LLE能够描述非线性流形的全局结构，例如那些人脸的数据集或者文本文档集-Manifold learning, embedded local linear algorithm (LLE), an intelligent algorithm to predict capture high-dimensional space contained in the low-dimensional features. And for local dimensionality reduction of the different clustering algorithms, LLE algorithm in a single low-dimensional characterization of the global coordinate system, the sampling space, but it does not optimize the local minimum. Reconstructed by the local symmetry of the linear study of the application, LLE can describe the global structure of nonlinear manifolds, such as those who face data sets or text document collection Platform: |
Size: 1024 |
Author: |
Hits:
Description: 问题规模化是近来信息学竞赛的一个新趋势，它意在通过扩大数据量来增加算法设计和编程实现的难度，这就向信息学竞赛的选手提出了更高层次的要求，本文试图探索一些解决此类问题的普遍性的策略。开始，本文给出了“规模化”一词的定义，并据此将其分为横向扩展和纵向扩展两种类型，分别进行论述。在探讨横向扩展问题的解决时本文是以谋划策略的“降维”思想为主要对象的；而重点讨论的是纵向扩展问题的解决，先提出了两种策略——分解法和精简法，然后结合一个具体例子研究“剪枝”在规模化问题中的应用。问题规模化是信息学竞赛向实际运用靠拢的一个体现，因此具有不可忽视的意义。-The scale of the problem is the recent informatics competition is a new trend, it intended to achieve by expanding the amount of data to increase the algorithm design and programming difficulty, which a higher level of requirements to the Olympiad players, this article attempts to explore some of thegeneral strategy to solve such problems. The paper gives the definition of the term "scale", which will be divided into horizontal expansion and vertical expansion of the two types, were discussed. Solution to explore the problem of horizontal expansion when this article is to plan strategy, "dimensionality reduction" ideology as the main object focus on the scale of problems, the first two strategies- decomposition and streamline the law, and then combined with a specific example of "pruning" in the large-scale problems. The scale of the problem is the the informatics race closer to the practical application of an embodiment and therefore can not be ignored significance. Platform: |
Size: 421888 |
Author:feiying |
Hits:
Description: 引入能直接处理连续型数据的邻域粗糙集约简模型，给出一种基于邻域粗糙集模型和粒子群优化的特征选择算法-Introduce neighborhood rough set reduction model can deal directly with continuous data, given the feature selection algorithm based on neighborhood rough set model and particle swarm optimization
Platform: |
Size: 13312 |
Author:伍洁 |
Hits:
Description: LLE算法是一种非线性数据降维方法，其特点是简单。-LLE algorithm is a kind of nonlinear data dimension reduction method.
Its characteristic is simple
Platform: |
Size: 2048 |
Author:huangyunjuan |
Hits:
Search - data reduction algorithm