取决于聚类成员资格的颜色图散点图
我对数据集进行软聚类,我想创建一个很酷的图形,看起来与发布的图像相似。我想以图形形式显示两个(或更多集群)之间的数据点成员资格。不过我不太确定该怎么做。我已经使用标准为数据点分配颜色,但不确定如何创建如下所示的更动态的图形。任何帮助表示赞赏。
Im conducting soft clustering on a data set and I wanted to create a cool graphic that looks similar to the image posted. I want to show a data points membership between two (or more clusters) in graphical form. Im not really sure how to go about this however. Ive used criteria to assign colours to a data point, but am unsure how to create a more dynamic sort of graphic seen below. Any help appreciated.
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。

绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(2)
我认为 标记 正是您要寻找的东西:
添加一些额外的代码来控制颜色和透明度(alpha)
I think markers are just the thing your looking for:
Adding someaditional code to control color and transparency (alpha)
scikit-learn
中的GaussianMixture
所做的事情与问题所要求的很接近。具体来说,
predict_proba(X)
返回一个数组,其中包含X
中每个点属于该分量的概率。在下面的示例中,我们拟合了两个混合组件,因此最后两个图应该彼此相反:The
GaussianMixture
inscikit-learn
does something close to what the question asks.Specifically,
predict_proba(X)
returns an array with the probability of each point inX
belonging to the component. In the example below we fit two mixture components, so the last two plots should be opposites of each other: