[riku_iki]

> However, it does not appear to have been peer reviewed.

my observation is that peer reviewers never try to reproduce results or do basic code audit to check that there is no data leak for example to training dataset.