machine learning
文章平均质量分 78
goooooooooo
怀胎十月生出个蛋,算了,接着孵
展开
-
regionprops [matlab]
regionprops:对labeled的matrix取属性 STATS = regionprops(L,properties)area:ROI的面积Centroid:质心 BoundingBox:包含ROI的最小矩形。返回值:[ul_corner width] ,ul_corner是upper left corner的坐标,width是矩形的长宽等(多维类推)MajorA翻译 2006-11-02 14:26:00 · 4588 阅读 · 2 评论 -
Graphical Models Links
概率图模型(Graphical Models)相关开源代码,如CRF,MRF,PLSA,LDA等,还有通用的有向和无向概率图相关工具原创 2011-05-23 13:43:00 · 1355 阅读 · 4 评论 -
杂记
>> massDet1(E:/work/change_classifier/训练集/1/PA1/ST1/SE1/);??? Error using ==> javaObjectNo class gradReader.BaggingGradExpandMLP can be located on Java class path. version(-java) 使用相同版本编译原创 2007-01-09 13:48:00 · 601 阅读 · 0 评论 -
function D=genFrctalDimension(img)
function D=genFrctalDimension(img)%get the fractal dimension of img[m,n]=size(img);Ur1=img;Br1=img;Ur=img;Br=img;count=100;arrayV=zeros(1,count);arrayR=zeros(1,count);for point=1:count for i=1:m原创 2007-01-03 13:31:00 · 693 阅读 · 0 评论 -
function writelog(filename,content)
function writelog(filename,content)%write log :write content to the file with name of filename%input filename:the filename of the log file%input content:the content of the log for writingdlmwrite(file原创 2007-01-03 13:27:00 · 758 阅读 · 0 评论 -
function return_value=EndWith(str1,str2)
function return_value=EndWith(str1,str2)%if str2 is the end of str1 then return 1 else return 0[tmp,len_1]=size(str1);[tmp,len_2]=size(str2);if len_1 return_value=0; return;endstart=len_1-len_2+原创 2007-01-03 13:33:00 · 709 阅读 · 0 评论 -
function returnFiles=getFiles(baseDir)
function returnFiles=getFiles(baseDir)% get files in a directory%input baseDir: is the directory for get Files,it must be a string end with%/%output returnFiles:is a cell of file paths in the direct原创 2007-01-03 13:27:00 · 713 阅读 · 0 评论 -
function bmp=convert2BMP(img)
function bmp=convert2BMP(img)%convert an image of 4096 gray levels to an img of 256 gray levels%input: img is of 4096 gray levels%output: bmp is of 256 gray levelsimg=double(img);bmp=uint8(floor(img/1原创 2007-01-03 13:28:00 · 785 阅读 · 1 评论 -
鸟话
Ensemble Learning : 三个臭皮匠,顶个诸葛亮原创 2007-01-02 12:57:00 · 788 阅读 · 0 评论 -
metacost用bagging实现cost-sensitive
This classifier should produce similar results to one created by passing the base learner to Bagging, which is in turn passed to a CostSensitiveClassifier operating on minimum expected cost. The dif转载 2006-12-18 10:50:00 · 4222 阅读 · 0 评论 -
ROC相关
终于搜到ROC的相关东东了,虽然没有具体讲FROC的ROC曲线receiver operating characteristic curve这个名称是评价雷达性能时采用的,称受信者操作特性曲线。即纵轴表示敏感度(真阳性率),横轴表示假阳性率=1-特异性,变化界限值时的两者的变化成一曲线。这ROC曲线中描写着可选择之任意的界限值中的敏感度和特异性。所以制成ROC图就能很容易查出,任意界限值中的病转载 2006-11-12 16:43:00 · 3147 阅读 · 0 评论 -
图像标注/分类相关
图像分割,特征提取,分类原创 2011-05-23 14:21:00 · 2565 阅读 · 0 评论