- 0
- 0
- 约3.58千字
- 约 11页
- 2026-09-30 发布于四川
- 举报
SeisClass:地震属性优化
2什么是Seisclass?基于地震属性之间的差异性或者相似性进行地震相划分的工具.优化/筛选地震属性,为后期油藏属性的预测做准备
3优化方法非督导:K-MeansClusteringCompetitivelearning督导:Fisher’sDiscriminantBayesianBackPropagation
4K-MeansClustering原理:TheK-MeansClusteringalgorithmisastatisticalmethod,andstartsbyrandomlyselectingclustercentersinattributespace–oneforeachclassthatwillbegenerated.Theuserpickthenumberofclasses.Asthealgorithmruns,alldatapointsareassignedtooneoftheclustercentersbasedonwhichclustercenterthedatapointisclosesttothemeanvalueofthecluster.Iterationscontinue
原创力文档

文档评论(0)