Exploring MLX

Exploring MLX Framework by Apple and comparing the performance with PyTorch on M1 Mac Air. To perform these experiments we used the MNIST Dataset and implemented a simple CNN architecture.

File Info:

• mnist.py contains the code for loading MNIST dataset in the MLX framework.
• mnist_in_mlx.py contains the main implementation of the CNN network in the MLX Framework.
• mnist_in_pytorch.py contains the implementation of the same network in PyTorch Framework.
• experiments.py is just a filler file used while coding to test out the various functions.