diff options
| author | Lorenzo Torres <lorenzotorres@outlook.it> | 2025-10-19 14:49:45 +0200 |
|---|---|---|
| committer | Lorenzo Torres <lorenzotorres@outlook.it> | 2025-10-19 14:49:45 +0200 |
| commit | 1e817b3c984c10c14b1e2d1623524f4d283bc1fe (patch) | |
| tree | 62e24281594ce411fb9af4e1c276420d672878b0 /content/projects | |
| parent | b1450631786e915e3ed365dfad18478e37c4ee84 (diff) | |
added MNIST to projects
Diffstat (limited to 'content/projects')
| -rw-r--r-- | content/projects/mnist.md | 12 |
1 files changed, 12 insertions, 0 deletions
diff --git a/content/projects/mnist.md b/content/projects/mnist.md new file mode 100644 index 0000000..e1bcdd8 --- /dev/null +++ b/content/projects/mnist.md @@ -0,0 +1,12 @@ +<!-- title: mnist --> + +## mnist +**mnist** is a neural network trained to predict images in the [MNIST database](https://en.wikipedia.org/wiki/MNIST_database). + +#### Description +It's written in python without making use of ML libraries directly, using Apple's MLX to accelerate matrix computations with the GPU. This makes the scripts very fast on Apple Silicon machines since MLX makes great use of the SoC shared memory and lazy evaluation for efficiency. The code is commented in detail and quite simple to understand with basic understanding of neural networks. + +#### Sources +mnist is currently hosted on GitHub at [https://github.com/lolzdev/mnist](https://github.com/lolzdev/mnist) + +#### Related posts |