Standard Projections
The following grid shows several dimensionality reduction techniques applied to the classic Iris dataset. Each method attempts to project the 4D data into 2D while preserving different aspects of the original structure.
PCA
Calculating...
MDS
Calculating...
t-SNE
Calculating...
UMAP
Calculating...
TriMap
Calculating...
ISOMAP
Calculating...
TopoMap
Calculating...
Sammon
Calculating...
LLE
Calculating...
FastMap
Calculating...
SMACOF
Calculating...
LDA
Calculating...
LSP
Calculating...
LTSA
Calculating...
SQDMDS
Calculating...
How-to (Code)
Most dimensionality reduction algorithms in DruidJS follow a standard interface. You instantiate the algorithm with your data and calling transform() to get the projection. You can find out more in the API Reference.
javascript
import * as druid from "@saehrimnir/druidjs";
const data = [
/* ... multi-dimensional data ... */
];
// Example: PCA
const pca = new druid.PCA(data);
const projectionPCA = pca.transform();
// Example: t-SNE (Iterative)
const tsne = new druid.TSNE(data, { perplexity: 30 });
const projectionTSNE = tsne.transform(1000); // 1000 iterations