Big Data Mapping
Clustering statistics box-plot.
ptSNE cost & size plot.
Class density maps plot.
Class density maps
Example dataset
Default bdm file name
Create bdm instance
Get data-point clustering labels.
Set/get default local machine name or IP address
Merging of clusters based on signal-to-noise-ratio.
Set/get default path for mybdm
Plots the signal-to-nois-ratio as a function of the number of clusters...
Find optimal number of clusters based on signal-to-noise-ratio.
Plot paKDE (density landscape)
Perplexity-adaptive kernel density estimation
Plot ptSNE (low-dimensional embedding)
Parallelized t-SNE
ptSNE quantile-maps
Save bdm instance
Transfer bdm instance to a remote machine.
Plot WTT (clustering)
Watertrack transform (WTT)
Unsupervised clustering protocol for large scale structured data, based on a low dimensional representation of the data. Dimensionality reduction is performed using a parallelized implementation of the t-Stochastic Neighboring Embedding algorithm (Garriga J. and Bartumeus F. (2018), <arXiv:1812.09869>).