08 2021 档案

matlab PCA分析
摘要:pca( ) 采用matlab自带的函数pca()进行主成分分析 [coeff, score, latent, tsquared, explained, mu] = pca(x) 假设数据x为n行p列的多变量数据,n为观测次数,p为变量维度。 coeff:为PCA变换系数,也称为loadings。 阅读全文

posted @ 2021-08-31 23:10 那抹阳光1994 阅读(5006) 评论(0) 推荐(0) 编辑

matlab计算马氏平方距离
摘要:马氏距离计算一个点到一个参考分布的距离。马氏距离的优点是不受各个维度特征尺度的影响(尺度无关)。 d2 = mahal(Y, X) d2 = mahal(Y,X) returns the squared Mahalanobis distance of each observation in Y to 阅读全文

posted @ 2021-08-31 10:13 那抹阳光1994 阅读(918) 评论(0) 推荐(0) 编辑

matlab计算排列组合
摘要:perms(x) perms(x) 给出向量x的所有排列。 perms(1:3) ans = 3 2 1 3 1 2 2 3 1 2 1 3 1 3 2 1 2 3 nchoosek(x,m) nchoosek(x,m)给出了从包含n个元素的向量x中选取m个元素的组合。 nchoosek(1:4,3 阅读全文

posted @ 2021-08-29 15:47 那抹阳光1994 阅读(636) 评论(0) 推荐(0) 编辑

matlab函数tight_subplot控制图像的边界(margin),subplot的间距(gap)
摘要:tight_subplot实现比matlab自带的subplot更紧凑的子图绘制。 function [ha, pos] = tight_subplot(Nh, Nw, gap, marg_h, marg_w) % tight_subplot creates "subplot" axes with 阅读全文

posted @ 2021-08-26 22:43 那抹阳光1994 阅读(2402) 评论(0) 推荐(0) 编辑

matlab获取数据直方图和经验累计分布CDF
摘要:histogram获取数据直方图 通过BinWidth参数调整箱子的宽度 BinWidth = 0.01 BinWidth = 0.1 cdfplot获取经验累积分布 阅读全文

posted @ 2021-08-26 17:44 那抹阳光1994 阅读(882) 评论(0) 推荐(0) 编辑

matlab计算数据分位点prctile
摘要:prctile计算数据集的百分位数(分位点) Y = prctile(X,p) 根据区间 [0,100] 中的百分比 p 返回数据向量或数组 X 中元素的百分位数。 a = rand(1, 10); sort(a) prctile(a, 70) a = 0.3452 0.2825 0.7467 0. 阅读全文

posted @ 2021-08-26 17:33 那抹阳光1994 阅读(4005) 评论(0) 推荐(0) 编辑

matlab散点矩阵图
摘要:1. gplotmatrix Matrix of scatter plots by group 按组划分的散点图矩阵 gplotmatrix(x,[],group) 创建 x 中数据的散点图矩阵,按 group 中的分组变量分组。 结果图中的每一组单独的轴都包含一个 x 列与另一列 x 的散点图。 阅读全文

posted @ 2021-08-25 22:44 那抹阳光1994 阅读(1658) 评论(0) 推荐(0) 编辑

Unicode字符表
摘要:常用的字符表,写论文可以用到 https://unicode-table.com/en/#basic-latin 阅读全文

posted @ 2021-08-17 17:13 那抹阳光1994 阅读(135) 评论(0) 推荐(0) 编辑

There is和There are怎么用?
摘要:There is/There are称为There be句型,也称为存在句型,表示事物的客观存在,There在此结构中是引导词,已经没有副词"那里"的含义 中文:在一个房间里有一个桌子,两个椅子。 翻译:There is a table and two chairs in the room. 例句: 阅读全文

posted @ 2021-08-14 17:37 那抹阳光1994 阅读(5186) 评论(0) 推荐(0) 编辑

a large amount of 和 large amounts of的区别?
摘要:a large amount of 和 large amounts of的区别: 1、a large amount of +不可数名词,谓语动词用单数;large amounts of+ 不可数名词, 谓语动词用复数 例:A large amount of money was spent on th 阅读全文

posted @ 2021-08-14 17:35 那抹阳光1994 阅读(4711) 评论(0) 推荐(0) 编辑

a lot,a large amount of,a large number of区别
摘要:a lot 修饰可数和不可数名词 a large amount of修饰不可数名词 a large number of修饰可数 阅读全文

posted @ 2021-08-14 17:33 那抹阳光1994 阅读(932) 评论(0) 推荐(0) 编辑

PyTorch中Tensor的维度变换
摘要:对于 PyTorch 的基本数据对象 Tensor (张量),在处理问题时,需要经常改变数据的维度,以便于后期的计算和进一步处理,本文旨在列举一些维度变换的方法并举例,方便大家查看。 维度查看:torch.Tensor.size() 查看当前 tensor 的维度 举个例子: >>> import 阅读全文

posted @ 2021-08-12 15:18 那抹阳光1994 阅读(864) 评论(0) 推荐(0) 编辑

pytorch 生成单位张量torch.eye
摘要:torch.eye(n, m=None, *, out=None, dtype=None, layout=torch.strided, device=None, requires_grad=False) → Tensor Returns a 2-D tensor with ones on the d 阅读全文

posted @ 2021-08-11 09:11 那抹阳光1994 阅读(414) 评论(0) 推荐(0) 编辑

pytorch Tensor.expand()张量扩张
摘要:Tensor.expand(*sizes) → 张量 返回自张量的新视图,单例维度扩展到更大的尺寸。 传递 -1 作为维度的大小意味着不更改该维度的大小。 Tensor 也可以扩展到更多的维度,新的维度会附加在前面。 对于新维度,大小不能设置为 -1。 扩展张量不会分配新的内存,而只会在现有张量上创 阅读全文

posted @ 2021-08-11 09:09 那抹阳光1994 阅读(5592) 评论(0) 推荐(0) 编辑

皮尔逊相关系数和余弦相似性的关系
摘要:皮尔逊相关系数和余弦相似性的关系 先上结论:在数据标准化( )后,Pearson相关性系数、Cosine相似度、欧式距离的平方可认为是等价的。详细推导在这里。 如何理解皮尔逊相关系数(Pearson Correlation Coefficient)? - 微调的回答 - 知乎 有两种理解方式 (1) 阅读全文

posted @ 2021-08-10 11:03 那抹阳光1994 阅读(1303) 评论(0) 推荐(0) 编辑

OAPI for Sap2000
摘要:通过OAPI可以调用Sap2000,实现参数化分析。 阅读全文

posted @ 2021-08-03 23:18 那抹阳光1994 阅读(138) 评论(0) 推荐(0) 编辑

PCA主成分的选择标准
摘要:采用PCA进行降维时,主成分的选择相当主观,这篇文章讨论了一些可用的方案: Selection of the Number of Principal Components: The Variance of the Reconstruction Error Criterion with a Compa 阅读全文

posted @ 2021-08-02 16:46 那抹阳光1994 阅读(239) 评论(0) 推荐(0) 编辑

序贯概率比检验Sequential Probability Ratio Test
摘要:序贯概率比检验(Sequential probability ratio test,SPRT) 什么是序贯概率比检验 数理统计学的一个分支,其名称源出于亚伯拉罕·瓦尔德在1947年发表的一本同名著作,它研究的对象是所谓“序贯抽样方案”,及如何用这种抽样方案得到的样本去作统计推断。序贯抽样方案是指在抽 阅读全文

posted @ 2021-08-02 10:54 那抹阳光1994 阅读(5342) 评论(0) 推荐(0) 编辑

导航

< 2025年3月 >
23 24 25 26 27 28 1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31 1 2 3 4 5
点击右上角即可分享
微信分享提示