# \[ANN\] UMAP.jl, a pure Julia implementation of Uniform Manifold Approximation and Projection

**URL:** <https://discourse.julialang.org/t/ann-umap-jl-a-pure-julia-implementation-of-uniform-manifold-approximation-and-projection/19829>\
**Category:** Package Announcements\
**Tags:** package, announcement\
**Created:** [January 19, 2019, 10:48pm UTC](https://discourse.julialang.org/t/ann-umap-jl-a-pure-julia-implementation-of-uniform-manifold-approximation-and-projection/19829 "2019-01-19T22:48:22Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![dillondaudert](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/dillondaudert/32/6795_2.png) [@dillondaudert](https://discourse.julialang.org/u/dillondaudert)\
**Post date:** [January 19, 2019, 10:48pm UTC](https://discourse.julialang.org/t/ann-umap-jl-a-pure-julia-implementation-of-uniform-manifold-approximation-and-projection/19829/1 "2019-01-19T22:48:22Z")

</div>

[UMAP.jl](https://github.com/dillondaudert/UMAP.jl) is a pure Julia implementation of the UMAP dimension reduction algorithm. From the abstract:

> UMAP (Uniform Manifold Approximation and Projection) is a novel manifold learning technique for dimension reduction. UMAP is constructed from a theoretical framework based in Riemannian geometry and algebraic topology. The result is a practical scalable algorithm that applies to real world data.

The first release, v0.1.0, is tagged in the General registry and can be installed with `]add UMAP`. More usage details are available in the README.

Example:

 ![MNIST UMAP.jl](https://global.discourse-cdn.com/julialang/original/3X/f/8/f8a4940d7401dd8eda329186a77555b2c0b51741.png)

---

<div class="post-metadata">

**Author:** ![Forbu](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/forbu/32/10728_2.png) [@Forbu](https://discourse.julialang.org/u/Forbu)\
**Post date:** [June 30, 2019, 9:47pm UTC](https://discourse.julialang.org/t/ann-umap-jl-a-pure-julia-implementation-of-uniform-manifold-approximation-and-projection/19829/2 "2019-06-30T21:47:55Z")

</div>

Good job !

Did you compare the speed of the algorithm between the python implementation of the original author (with numba speed up) and your version ?

I want to see which version I have to use 🙂