Evaluation Metrics for the default model weights #113
Description
In addition to the Benchmark info provided in the README of this repo, it will be highly helpful and good to know what was the training and evaluation loss / accuracy for the default model weights.