/computer_vision

notebooks exploring different vision techniques and gradually building a from-scratch PyTorch vision library

Primary LanguageJupyter Notebook

Computer Vision

In this repository, I implement computer vision techniques and ideas that are initially new to me in notebooks. Then, if the notebook contains blocks of code that can be re-used for different vision tasks, I put them in files under the src directory.

In this iterative fashion, I'm building a mini vision library built off of pure PyTorch.

This repository goes from Lenet-5, to diffusion.

Notebooks

  • mnist_lenet.ipynb: training LeNet-5 on MNIST.
  • autoencoder.ipynb: autoencoders.
  • dcgan.ipynb: Deep convolutional Generative Adversarial Networks.
  • resnets.ipynb: residual networks.
  • fcn_Segmentation.ipynb: fully convolutional networks for semantic segmentation.
  • ddpm_mnist.ipynb: implementing diffusion models architecture on MNIST.
  • ddpm_cifar.ipynb: applying the DDPM architecture from src to CIFAR-10.
  • mixture_of_experts.ipynb: implementing a sparsely-gated mixture of experts with ResNets.