ML Basic #6 | Dimensionality Reduction

    2. Machine Learning  Comments

Introduction


1. UMAP

  • Introduction : a fairly flexible non-linear dimension reduction algorithm
  • Usage :
    • 2-dim representation of given high dimensional dataset
  • Parameters : Basic UMAP Parameters — umap 0.5 documentation
    • n_neighbors
    • min_dist: controls how tightly UMAP is allowed to pack points together. It provides the minimum distance apart that points are allowed to be in the low dimensional representation
    • n_components : allows the user to determine the dimensionality of the reduced dimension space we will be embedding the data into.
    • metric : controls how distance is computed in the ambient space of the input data (euclidean, cosine, hamming, etc.)

2. Sub-TItle

  • docs

3. Sub-TItle

  • docs