Skip to content

A simple MATLAB wrapper for umap. Supports un/semi/supervised embedding.

Notifications You must be signed in to change notification settings

sg-s/umap-matlab-wrapper

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

22 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

MATLAB wrapper for umap

This repo provides a simple wrapper for umap

Installation

  1. Install a required dependency using git clone [email protected]:sg-s/srinivas.gs_mtools.git
  2. Add that folder to your MATLAB path. Don't add subfolders.
  3. Install umap using conda install -c conda-forge umap-learn (paste this in your shell, not your MATLAB prompt)
  4. Install h5py using conda install h5py
  5. (optional) Install condalab
  6. (optional) Run conda.init to configure condalab in MATLAB

Use

u = umap();
R = u.fit(X);

Parameters and options

Many of the parameters and options in umap are exposed in the object, and you can change these directly from MATLAB. For example:

u = umap

  umap with properties:

             n_neighbors: 15
            n_components: 2
                  metric: 'euclidean'
           learning_rate: 1
                min_dist: 0.1000
                  spread: 1
        set_op_mix_ratio: 1
      local_connectivity: 1
      repulsion_strength: 1
    negative_sample_rate: 5
    transform_queue_size: 4
      target_n_neighbors: -1
           target_weight: 0.5000
          transform_seed: 42


u.n_neighbors = 10;
u.metric = 'precomputed';

License

GPL

About

A simple MATLAB wrapper for umap. Supports un/semi/supervised embedding.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published