如何在python中可视化/绘制SOM?

如何在python中可视化/绘制SOM?,python,visualization,som,self-organizing-maps,Python,Visualization,Som,Self Organizing Maps,我想为我的集群结果创建一个映射。例如,这是我使用SOM生成的结果 我的输入示例(基于DNA序列基序): 基序1=0.19,0.95,0.01,0,0.76,0,1.04,0,0,0.05,0,1,0,0,0,1,0 基序2=0,0,0,0,0,0,1,0.3,0.05,0.15,0.7,0.6,0.05,1.15,0.2,0.8 母题3=0.9,0,0,1.1,0,0,0,0,0.45,0.035,0,0.015,0.15,1.665,0,0.335,1.35 模体4=1,0,0,1.16,0.

我想为我的集群结果创建一个映射。例如,这是我使用SOM生成的结果

我的输入示例(基于DNA序列基序):

基序1=0.19,0.95,0.01,0,0.76,0,1.04,0,0,0.05,0,1,0,0,0,1,0

基序2=0,0,0,0,0,0,1,0.3,0.05,0.15,0.7,0.6,0.05,1.15,0.2,0.8

母题3=0.9,0,0,1.1,0,0,0,0,0.45,0.035,0,0.015,0.15,1.665,0,0.335,1.35

模体4=1,0,0,1.16,0.036,0,0.0032,0.4,0.294,0,0.025,0.04,1.5888,0.04,0.371,1.04

输出(在python中使用SOM运行): 培训投入集群: 基序1=簇1

基序2=簇2

基序3=簇1

基序4=簇1

节点1的权重: 1.366,0.951,0.819,0.919,0.812,0.688,0.802,0.622,0.999,0.574,0.618,0.803,0.880,0.721,0.741,0.963 节点2的权重:
1.366,0.951,0.819,0.919,0.812,0.688,0.802,0.622,0.999,0.574,0.618,0.803,0.880,0.721,0.741,0.963

我一直在寻找同样的方法,但找到了一个好的替代方法。它被称为星暴图,在一个名为popsom的SOM实现中,由Lutz Hamel博士和他的前学生开发和维护


是否仅使用两个节点?绘制散点图或
plt.imshow
如何,其中每个节点表示二维网格上的一个点,并且颜色对应于该节点中的主要类别?在SUSI包中,这被称为
估计图
,这可能有助于: