View on GitHub

ipcv2022

Repository for IMPA's 2022 Image Processing and Computer Vision Course

Pytorch

For this lab, we recommend that readers watch the videos in the references section on this page. Let us work with a neural network for fashion classification that uses the FashionMNIST dataset. We have prepared an example notebook. To access it, click here.

Suggestions:

  1. Run the notebook cells and note all output and runtime.
  2. Change the number of epochs and check if the network accuracy improves.
  3. Change the device to GPU and rerun the notebook.
  4. If you are using the notebook on a local machine, you may need to make a minor modification to the code.
  5. Use the notebook to create a new network using one of the datasets available in the Pytorch webpage’s list. Click here to access it.

References

Introduction to Pytorch

Curso Online - Processamento de Imagens - Aula 05

Introduction to Neural Networks with Pytorch

Introdução ao PyTorch -  Redes Neurais - Primeiros passos com Pytorch - Deep Learning #1