Skip to main content

Dimensionality Reduction

We support three dimensionality reduction algorithms. The Number of Components parameter defines the desired dimensionality of the output data.

Accessing Dimensionality Reduction in evoML​

  1. Create a New Trial

  2. Navigate to Feature Engineering and then Feature Dimensionality Reduction dimensionality reduction

Dimensionality Reduction Algorithms​

1. Singular Value Decomposition​

A technique for decomposing a matrix into three separate matrices that capture the underlying structure of the original matrix.

2. Principal Component Analysis​

A technique for reducing the dimensionality of high-dimensional data by finding the most important patterns in the data and projecting it into a lower-dimensional space.

3. Independent Component Analysis​

A technique used for separating a multivariate signal into individual, non-gaussian components.