gradio_article.md

Training

Fast.ai was used to train this classifier with a ResNet50 vision learner, without much customization. The final fine tuning of the training loop resulted in the following losses. epoch train_loss valid_loss error_rate time 0 0.494005 0.078781 0.017094 00:05 epoch train_loss valid_loss error_rate time 0 0.089555 0.031788 0.025641 00:04 1 0.045144 0.097199 0.017094 00:03 2 0.033626 0.106634 0.017094 00:03

Examples

The example images provided in the demo are not from the dataset. They were quickly sourced via a couple of DuckDuckGo searches.