@burkej
Nothing here yet.
Nothing here yet.
Feb 10 · 1 min read · While working on our camera and file upload system, we were still not satisfied with the results our model was giving. We then decided to go back to our Google Colab notebook and really explore the training images. This is when we discovered that our...
Join discussionFeb 6 · 2 min read · In the previous post, we discussed choosing fashionMNIST as our training dataset, mainly due to its size and the nature of this application. We are going to use a Google Colab notebook for training so that multiple people can edit it, and we do not h...
Join discussionJan 29 · 1 min read · After further research of datasets and taking into account the nature of our application, we decided to re-explore fashionMNIST as a dataset we could use to train our model. Although it does not handle noise particularly well, from some ad hoc testin...
Join discussion