-
Notifications
You must be signed in to change notification settings - Fork 4
fansy1990/fast_cluster
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
--------------------------------------------------- 运行说明:(linux操作默认使用root用户) 1. 拷贝Jfreechar的相关包:jfreechar-1.0.15.jar , jcommon-1.0.17.jar 到$HADOOP_HOME/share/hadoop/mapreduce 2. 拷贝 digit_recoginzation_train_small.csv 到hdfs的/user/root/digit.csv文件 hadoop fs -put /opt/digit_recoginzation_train_small.csv /user/root/digit.csv 3. 云平台MR任务 3.0 把src/main/java目录的源码export到 fast_cluster.jar文件 3.1. 转换为序列化文件 hadoop jar /opt/fast_cluster.jar fz.fast_cluster.ToSeqJob digit_train.csv seq_out , 输出有5975条记录 3.2. 计算距离 hadoop jar /opt/fast_cluster.jar fz.fast_cluster.CalDistanceJob seq_out cal_distance_out 输出有17847325条记录 17847325 = 5975*(5975-1)/2 3.3. 根据距离求解阈值,并根据阈值画决策图 hadoop jar /opt/fast_cluster.jar fz.fast_cluster.CalDiffDCFindVec cal_distance_out/part-r-00000 17847325 0.02 0.02 1 2 2 /opt/picandlog 500 输出5570条记录 3.4. 根据距离和密度阈值求得聚类中心并写入本地 hadoop jar /opt/fast_cluster.jar fz.fast_cluster.FindWriteCenter /opt/picandlog/log_center_0.dat 500 1000 digit_train.csv center_out /opt/picandlog/center.dat 4. 根据聚类中心对每个数据进行分类 未完成!
About
No description, website, or topics provided.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published