umap-learn - Reduce high-dimensional data with UMAP
Apply UMAP for nonlinear dimensionality reduction, visualization, supervised embeddings, new-data transformation, and clustering preprocessing.
Tags
Updated: 2026-09-29Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Reduce data dimensionality
- Visualize low-dimensional embeddings
- Prepare clustering features
- Use supervised labels
- Transform new data
- Tune embedding parameters
Inputs
- High-dimensional data
- Feature labels
- New data
- UMAP parameters
- Clustering settings
Outputs
- Low-dimensional embeddings
- Transformed data embeddings
- Cluster labels
- Visualization plots
- Evaluation metrics
Requirements
- Python environment
- umap-learn package
- scikit-learn for preprocessing
- HDBSCAN for clustering
- Matplotlib for plotting